View
217
Download
1
Category
Preview:
Citation preview
REFERENCE CONDITIONS AND
EUTROPHICATION IMPACTS IN IRISH
RIVERS: MEETING THE
REQUIREMENTS OF THE WATER
FRAMEWORK DIRECTIVE
FINAL REPORT
(Project 2000-FS-2-M1)
Prepared for the Environmental Protection Agency
Author:
Aisling Walsh
ENVIRONMENTAL PROTECTION AGENCY
An Ghníomhaireacht um Chaomhnú Comhshaoil
PO Box 3000, Johnstown Castle, Co.Wexford, Ireland
© Environmental Protection Agency 2005
DISCLAIMER
Although every effort has been made to ensure the accuracy of the material contained
in this publication, complete accuracy cannot be guaranteed. Neither the
Environmental Protection Agency nor the author(s) accept any responsibility
whatsoever for loss or damage occasioned or claimed to have been occasioned, in part
or in full, as a consequence of any person acting, or refraining from acting, as a result
of a matter contained in this publication. All or part of this publication may be
reproduced without further permission, provided the source is acknowledged.
REFERENCE CONDITIONS AND EUTROPHICATION IMPACTS IN IRISH
RIVERS: MEETING THE REQUIREMENTS OF THE WATER FRAMEWORK
DIRECTIVE
FINAL REPORT
Published by the Environmental Protection Agency, Ireland
Printed on Recycled paper
ISBN:1-84095- to come 06/01/300
Price: €XX.XX
Abstract
An important objective of this project was to obtain an improved
understanding of why ecological communities depart from reference
conditions as pollution and eutrophication impacts on individual species and
in particular indicator taxa such as Ecdyonurus. The effects of
eutrophication on Ecdyonurus were studied using a novel split-stream
experiment which involved artificially increasing the phosphorus
concentrations in two oligotrophic rivers in the West of Ireland. Some of
the nutrient manipulation experiments showed significant differences in
algal biomass between the control and treated sections, but not all did so.
Findings from this study appear to indicate that Ecdyonurus is not directly
affected by the consequences of eutrophication.
The experiments did reveal surprising results showing the importance of N-
limitation in the rivers studied. On analysing the N:P ratios in a number of
rivers in the West of Ireland it was found that approximately 4% of the
samples were N-limited with low MRP concentrations (<0.05mg/l P). This
implies that a small proportion of high status rivers are N-limited rather than
P-limited. Thus, in terms of the Water Framework Directive, there may be a
case for introducing tighter regulations on the limits of nitrogen emitted to
water bodies as well as the need to control phosphorus. Results from these
studies highlight the complexity of the in-stream processes driving these N-
limited and P-limited high status rivers.
It was hypothesised that the changes exhibited in the biomass in the split-
stream experiment would be reflected in the gut contents of Ecdyonurus
venosus, possibly showing a variation in algal taxa between both sections of
the manipulation experiment. The split-stream experiment showed that
Ecdyonurus did not demonstrate distinct preferences due to enrichment but
neither were there any observed changes in the periphyton species
community. On the basis of the findings from the study in the Castlebar
River in 2003, Ecdyonurus venosus can be classified as both a herbivorous
grazer and detritus feeder with a tendency towards opportunistic feeding.
ii
The food preference of these larvae appears to be strongly dependent on the
food available in the environment at a given time and they seem to feed on
particles that are most abundant during a particular season or those that are
easily accessible during a feeding episode. The Ecydonurus gut contents
consisted mainly of epilithic algal tissue, plant particulate matter (detritus),
biofilm matrix and inorganic debris (mineral material).
The life cycle of the Heptegeniidae was described for the first time in detail
in five high status rivers in the West of Ireland. Of the three species of
Ecdyonurus studied, Ecdyonurus venosus was the dominant species in all
five high status rivers displaying a bivoltine life cycle with only a slight
variation in emergence periods between sites. Findings show that one
would expect to find this species when sampling during all seasons
throughout the year. The life cycle of both Ecdyonurus insignis and
Ecdyonurus dispar were found to be univoltine. The life cycle analyses in
this study suggest that at least one species of Ecdyonurus should be present
at all times of the year.
The life cycle of Rhithrogena semicolorata was substantially easier to
interpret and clearly displayed a univoltine life cycle. The Heptagenia
specimens were identified to genus level only and were therefore described
as a genus group that appeared to adopt a univoltine life cycle. Findings
from our studies support the hypothesis put forward that the various species
of Ecdyonurus emerge in overlapping phases such that during the summer
months larvae of at least one Ecdyonurus species will be present in the
benthic riffle fauna of Irish rivers
Ecdyonurus is a good indicator of pollution and the water chemistry results
appear to support the hypothesis that the presence of Ecdyonurus is
associated with good water quality. The presence/absence of Ecdyonurus in
the high status versus impacted sites supports its use as a significant
bioindicator of water quality. A selection of indices and metrics were
examined and the results revealed that the most significant differences
between the high status and impacted sites were found using Margalef’s
index, total number of taxa and % EPT. Results from feeding and
iii
microhabitat investigations suggest that as eutrophication and the impacts of
organic pollution progress, a change in the feeding guilds and microhabitat
preferences among the macroinvertebrate communities occurs.
iv
ACKNOWLEDGEMENTS
This report has been prepared as part of the Environmental Research
Technological Development and Innovation Programme under the
Productive Sector Operational Programme 2000-2006. The programme is
financed by the Irish Government under the National Development Plan
2000-2006. It is administered on behalf of the Department of the
Environment and Local Government by the Environmental Protection
Agency which has the statutory function of co-ordinating and promoting
environmental research.
My sincere thanks to my supervisor Dr. Mary Kelly-Quinn for her advice
and guidance throughout the course of this study. I would like to thank
Professors John Bracken and Tom Bolger for the use of departmental
facilities, and my committee members Dr. Bret Danilowicz and Dr. Tom
Hayden for their helpful suggestions and comments.
This project was funded by the Irish Environmental Protection Agency
under the ERDTI programme 2000-2006, their sponsorship is gratefully
appreciated. I wish to convey a special thanks to Mr. Martin McGarrigle,
EPA, Castlebar for his knowledge, assistance and advice during the course
of the last five years.
I would like to thank the chemistry laboratory in the EPA Regional
Inspectorate in Castlebar for processing the water samples, to Hugh
McGinley and Pat Durkin from the EPA Hydrometric Office, Castlebar, for
the help with constructing the split-stream experiment and to Noriana
Kennedy and Eli Mulvey for assistance with the laboratory and fieldwork.
Many thanks to Berine Ni Chathain for providing invaluable assistance with
diatom identifications and to Dr. Robin Raine in the Martin Ryan Marine
Institute in NUI, Galway for the use of a fluorescent microscope.
I would like to thank the staff and postgraduates in the Limnology unit in
UCDs’ Zoology Department, especially Wayne Trodd. The help provided
v
by Fiona Kelly for assistance with the sediment analysis is gratefully
acknowledged.
To my colleagues in the EPA in Castlebar, particularly Karol Donnelly and
Ruth Little for their help, good humour and encouragement over the years.
vi
TABLE OF CONTENTS
____________________________________________________________________
Title page i
Abstract ii
Acknowledgements v
Table of Contents vii
Chapter 1. Introduction 1 ____________________________________________________________________
1.1 Introduction 1
1.2 Thesis outline 4
1.3 Project context 6
1.4 The Water Framework Directive 9
1.4.1 Background 9
1.4.2 Ecological boundaries 9
1.4.3 The Reference condition 10
1.4.4 Classifying ecological status 12
1.4.5 Benthic macroinvertebrates 13
1.4.6 Conclusions 14
1.5 Eutrophication 15
1.5.1 Background 15
1.5.2 Nutrient sources 15
1.5.3 Eutrophication Trends in Ireland 16
1.5.4 Effect of eutrophication on macroinvertebrates 17
1.5.5 Diurnal dissolved (DO) variation 18
1.6 Research Objectives 20
1.7 Study locations 21
1.7.1 Site 1: The Owengarve River 24
1.7.2 Site 2: The Castlebar River 26
1.7.3 Site 3: The Brusna River 28
1.7.4 Site 4: The Dunneill River 30
vii
1.7.5 Site 5: Callow Loughs Stream 32
1.7.6 Site 6: The Cartron River 34
1.7.7 Site 7: Lough na Corralea Stream 36
1.7.8 Site 8: The Robe River 38
1.7.9 Site 9: The Mullaghanoe River 40
1.7.10 Site 10: The Mad River 42
Chapter 2. Experimental enrichment of a high status river in the
West of Ireland: Effects of nutrient manipulation on the genus
Ecdyonurus and benthic chlorophyll levels 44 _____________________________________________________________________
2.1 Introduction 44
2.2 Study outline 47
2.3 Materials and Methods 48
2.3.1 Site 1: The Clydagh River 48
2.3.2 Site 2: The Castlebar River 48
2.3.3 Preliminary investigations 49
2.3.4 Experimental design and nutrient addition 50
2.3.4.1 Split-stream construction 50
2.3.4.2 Nutrient addition 52
2.3.4.3 Chlorophyll a estimations 53
2.3.4.4 Macroinvertebrate sampling 56
2.4 Results 57
2.4.1 Effects of nutrient enrichment on periphyton growth 61
2.4.1.1 Experiment1-Clydagh River: 26th
July-15th
October 2002 61
2.4.1.2 Experiment 2-Clydagh River: 5th
June-11th
July 2003 62
2.4.1.3 Experiment 3-Clydagh River: 18th
July-12th
Septembe r 2003
63
2.4.2 Effects of nutrient enrichment on macroinvertebrates 64
2.4.2.1 Experiment 1-Clydagh River: 26th
July-15th
October 2002 64
2.4.2.2 Experiment 2-Clydagh River:5th
June-11th
July 2003 66
2.4.2.3 Experiment 3-Clydagh River: 18th
July-12th
September 2003
66
viii
2.5 Discussion 68
2.5.1 Introduction 68
2.5.2 General findings in the Clydagh River 2002 68
2.5.3 General findings in the Clydagh River 2003 70
2.5.4 General findings in the Castlebar River 2003 71
2.5.5 Comparisons with other studies 73
2.5.6 N:P Ratios 74
Chapter 3. Investigations into the feeding regime of the mayfly
Ecdyonurus venous 76 _____________________________________________________________________
3.1 Introduction 76
3.2 Study outline 83
3.3 Materials and Methods 84
3.3.1 Site description: The Castlebar River 84
3.3.2 Sampling design 84
3.3.3 Gut content analysis 85
3.3.3.1 Cold acid hydrogen peroxide method 85
3.3.3.2 DAPI 86
3.3.4 Analysis Benthic stone scrapings 89
3.4 Results 91
3.4.1 Gut content analysis 91
3.4.1.1 Gut content results ‘pre-nutrient addition’-18th
July 2003 93
3.4.1.2 Gut content results ‘post-nutrient addition’-12th
September
2003 97
3.4.1.3 Gut content results from additional samples investigated 100
3.4.1.4 Results of gut analysis of Ecdyonurus venosus during spring
2003 investigation 106
3.4.2 Analysis of stone scrapings 112
3.4.2.1 Stone scraping results ‘pre-nutrient addition’ 21st July 2003
112
3.4.2.2 Stone scraping results ‘post-nutrient addition (9th
September
2003) 114
ix
3.4.2.3 Algal composition in pre- and post-nutrient addition samples
117
3.4.2.4 Summary of results from epilithic scrapings 123
3.4.3 Comparison of Ecdyonurus Gut Contents with Stone Scrapings 125
3.5 Discussion 130
3.5.1 Introduction 130
3.5.2 Summary of findings 131
Chapter 4. Life history studies of the Heptageniidae family in five
high status rivers in the West of Ireland 137 _____________________________________________________________________
4.1 Introduction 137
4.2 Study outline 143
4.3 Materials and Methods 144
4.3.1 Experimental design and sampling regime 144
4.3.2 Laboratory studies: Identification and categorisation of monthly
samples 145
4.3.3 Rearing experiments 148
4.3.3.1 Rearing chamber 148
4.3.3.2 Data collection 148
4.3.3.3 Laboratory rearing 149
4.4 Results
4.4.1 Physico-chemical data 151
4.4.2 Results from life cycle studies 151
4.4.2.1 Interpretation of the life cycle of Ecdyonurus venosus 152
4.4.2.2 Interpretation of the life cycle of Ecdyonurus insignis 159
4.4.2.3 Interpretation of the life cycle of Ecdyonurus dispar 162
4.4.2.4 Interpretation of the life cycle of Rhithrogena semicolorata 166
4.4.2.5 Interpretation of the life cycle of Heptagenia species 172
4.4.2.6 Temperature variation among seasons 181
4.4.3 Results from the rearing experiments 183
4.5 Discussion 185
4.5.1 Introduction 185
x
4.5.2 Interpretation of the life history of the genus Ecdyonurus 186
4.5.3 Interpretation of the life cycle of Rhithrogena semicolorata 196
4.5.4 Interpretation of the life cycle of Heptagenia species 199
4.5.5 Rearing trials 200
4.5.6 Concluding comments 201
Chapter 5. Assessment of the physical, chemical and biological
factors controlling the occurrence of Ecdyonurus in five high status
and five impacted rivers in the West of Ireland 203 _____________________________________________________________________
5.1 Introduction 203
5.2 Study outline 208
5.3 Materials and Methods 209
5.3.1 Preliminary site investigations 209
5.3.2 Physico-chemical parameters - procedures and analysis 211
5.3.3 Macroinvertebrate collection 212
5.3.3.1 Sampling procedure 212
5.3.3.2 Determination of optimum numbers of Surber samples 214
5.3.3.3 Analysis of sample data 216
5.3.4 Sediment analysis 221
5.3.4.1 Sampling procedure 221
5.3.4.2 Sample Treatment 221
5.3.4.3 Analysis of samples 222
5.3.4.4 Data presentation 223
5.3.5 Night-time Dissolved Oxygen (DO) measurements 223
5.4 Results 224
5.4.1 Physico-chemical results 224
5.4.1.1 Temperature (°C) and pH 225
5.4.1.2 Conductivity (µS/cm) and alkalinity (mg/l CaC03 ) 226
5.4.1.3 Dissolved Oxygen (% saturation) and BOD (mg/l 02 ) 229
5.4.1.4 Colour (Hazen units) and Chloride (mg/l Cl) 230
5.4.1.5 Ammonia (mg/l N) 232
5.4.1.6 TON (mg/l N) 233
xi
5.4.1.7 Unfiltered MRP (mg/l P) 234
5.4.1.8 Statistical comparisons of physico-chemical results between the
high status v impacted sites 235
5.4.1.9 Investigation in the NP ratios between high status v impacted
sites 237
5.4.2 Night time dissolved oxygen (DO) measurements 242
5.4.3 Results of the macroinvertebrate analysis 243
5.4.3.1 Shannon-Wiener Index (H’) 244
5.4.3.2 Margalef’s Index (d) 245
5.4.3.3 Pielous’s eveness (J’) 246
5.4.3.4 Simpsons diversity 247
5.4.3.5 Total number of taxa (S) 248
5.4.3.6 Taxon abundance (N) 249
5.4.3.7 Percentage EPT 250
5.4.3.8 Seasonal variations 251
5.4.3.9 Calculation of AQEM metrics 253
5.4.3.10 Microhabitat preferences 255
5.4.3.11 Feeding types 257
5.4.4 Results of sediment analysis 258
5.4.4.1 Substrate fractions – High status rivers 258
5.4.4.2 Substrate fractions – Impacted sites 261
5.4.4.3 Cumulative frequency curves 264
5.4.4.4 Fine sediment (particles < 1mm diameter) 265
5.5 Discussion 269
5.5.1 Introduction 269
5.5.2 General discussion 269
Chapter 6. General Discussion 276 _____________________________________________________________________
6.1 General discussion 276
6.1.1 Split-stream experiment 276
6.1.1.1 Comments and recommendations 278
xii
6.1.2 Investigation into the feeding regime of Ecdyonurus venosus
279
6.1.2.1 Comments and recommendations 281
6.1.3 Life history studies of the Heptageniidae in five high status
rivers in the West of Ireland 282
6.1.3.1 Comments and recommendations 284
6.1.4 Examination of the biotic and abiotic factors controlling the
distribution of the genus Ecdyonurus 286
6.1.4.1 Comments and recommendations 290
6.1.5 Concluding comments 291
References 293
Appendices 325
xiii
1
Chapter 1
Introduction
1.1 Introduction
A fundamental goal of ecology is to synthesise information about the natural
world and then to predict how the structure and function of species,
populations, communities and/or ecosystems respond to change. There is
particular interest among lotic ecologists to find ways of predicting the
effects of change on freshwater systems, not only because streams are
naturally variable, but also because streams and rivers are vulnerable to
anthropogenic disturbances (Power et al., 1988). Water, once a seemingly
unlimited renewable resource, is in direct need of protection and
conservation as aquatic environments are now among our most endangered
habitats. The restoration and maintenance of healthy river ecosystems have
become important objectives of river management and monitoring
programmes (Gore, 1985; Karr, 1991; Rapport, 1991). The overall aim of
monitoring programmes is to assess ecological communities and to
determine the environmental factors that are responsible for altering them.
The ecological community, however, is among the most difficult entities to
measure, define and understand. So, while the concept is attractive, its
complexity confounds attempts of unique and concise measurements that
account for the environmental factors or disturbances directly responsible
for deleterious alterations (Moog, 1995).
This problem can be overcome by adapting some species level concepts.
Many physical characteristics of the environment have direct influence on
and to a great degree determine the distribution, abundance and behaviour
of individual organisms and the populations to which they belong. The
study of how an individual species responds to its environment, or
2
moreover, specific characteristics of the environment, is termed autecology
(Moog, 1995). Environmental pressures of anthropogenic origin can disrupt
the ecological integrity of a system causing a change in autecological
relations.
The complexity of these interactions in lotic ecosystems is not yet fully
understood. Factors that are particularly important include the influence of
local physical conditions on organisms, life histories and the effects of
species interactions on lotic communities and ecosystems. The relative
importance of local and contemporary versus biogeographic or historical
factors in determining species distribution and the response in lotic
communities to major environmental changes, needs further (Power et al.,
1988).
Mechanisms, responses and interactions between organisms and the
environment are so numerous that it is impractical to measure but a few.
Bioindicators play a very important role in measuring the effects of
environmental stress in aquatic habitats. Bioindicators are species that
interact with key environmental factors resulting in observable and
characteristic responses that alter both the makeup and the ecological
function of a particular community. The use of species as opposed to higher
taxonomic groups, is emphasised by Moog (1995) in his studies, and he and
his co-workers strongly agree that most higher taxonomic groups are either
too heterogeneous in their response to environmental stress or lack the
useful sensitivity that individual species have to environmental factors.
Relationships between biotic and abiotic factors and the composition and
structure of lotic macroinvertebrate communities have received considerable
attention (Clenaghan et al., 1998). Temporal changes in macroinvertebrate
composition have been related to life history patterns in the community
(Egglishaw and Mackay, 1967; Armitage et al., 1975), which have been
hypothesised to have evolved in response to food availability (e.g. Vannote
et al., 1980) and seasonal changes in physico-chemical factors (e.g. Bunn et
al., 1986). The temporal and spatial distribution of an organism is tightly
connected to its physiological responses to varying levels of environmental
3
factors. If these responses are known, then the distribution of an organism
can be used to indicate the magnitude of these environmental factors. The
key biotic and abiotic factors that influence aquatic organisms are water
temperature and oxygen balance (Horner and Welch, 1981), food (nutrient)
composition and availability (McCormick and Stevenson, 1991;
Mullholland et al., 1991; Hill, Boston and Steinman, 1992), light intensity
(Steinman et al., 1989), current-substrate relation (Richards et al., 1993) and
habitat structure or cover (Hawkins et al., 1982). Autecological relations,
that is, the interactions between individuals and these environmental factors
are obviously the fundamental elements in discovering the causes of
community disturbance.
4
1.2 Thesis outline
This investigation focuses on an autecological study of the genus
Ecdyonurus in relation to some of the key biotic and abiotic factors
controlling its distribution, in five high status and five impacted rivers in the
West of Ireland. The project was divided into four key areas, each
investigating a particular aspect of the overall ecology of Ecdyonurus.
Chapter 2 investigates the effect of artificial eutrophication on an
oligotrophic system. Little is know about the effects of eutrophication on
Ecdyonurus. The aim was to artificially increase the concentration of
phosphorus and observe the response of Ecdyonurus using a novel split-
stream devised specifically for this project.
To date, there have been no investigations in Ireland into the diet of
Ecdyonurus and this project includes a study of the feeding ecology of this
macroinvertebrate. Chapter 3 examines and describes the gut contents of
Ecdyonurus venosus (Fabricius) in order to establish the preferred diet of
this mayfly. The larvae were taken from the Castlebar River, the same river
that was used for the split-stream experiment outlined in Chapter 2. A
preliminary investigation was carried out on Ecdyonurus specimens taken
from a number of the high status rivers.
More detailed knowledge of the life history of Ecdyonurus is required in
order to improve the sensitivity of the EPA Q-value System and also to
interpret the impact of pressures on the ecological status of rivers. Chapter
4 outlines the detailed investigation into the life cycle of the four species of
Ecdyonurus and other Heptageniidae species studied in the five high status
rivers selected for this project.
At the high status sites, Ecdyonurus are able to flourish and maintain
sustainable populations while at the impacted sites these taxa once survived
but for reasons unknown, no longer exist. Chapter 5 outlines the
comprehensive sampling programme undertaken in the five high status and
five impacted rivers over a 16-18-month period which included monthly
5
macroinvertebrate and river water sampling. The riverbed substrates were
also sampled in order to establish the sediment fraction sizes in each of the
ten sites. A range of potential biotic interactions between Ecdyonurus and
other members of the faunal community was also investigated, comparing
the faunal communities at the impacted sites with those at the high status
sites. The purpose of this investigation was to increase our knowledge of
the changes that occur during the eutrophication process by studying and
comparing the differences in the chemical, physical and biotic factors
between the high status and impacted sites.
6
1.3 Project context
The genus Ecdyonurus is a key water quality bioindicator and its presence
or absence has been used as an important element of the EPA’s biological
River Quality System (Q-value System) since 1970 for the ecological
assessment of Irish rivers (Flanagan and Toner, 1972; McGarrigle et al.,
2002). This system has proved to be very successful in defining water
quality in rivers. Table 1.1 illustrates the relationship between the Irish Q-
value System and traditional chemical measures of organic pollution and
eutrophication (adapted from McGarrigle, 2001). As Q-value improves the
concentrations of biochemical oxygen demand (BOD), ammonia, nitrate,
and phosphate decline significantly. The Q-value System is therefore
clearly linked to standard chemical measures of water quality.
Table 1.1 'Typical' physico-chemical values for Q-value categories: median,
mean and standard deviation of the nationally reported values for all sites
within each Q-value band. (Adapted from McGarrigle, 2001).
Q-value
Q1
Q1-2
Q2
Q2-3
Q3
Q3-4
Q4
Q4-5
Q5
Max No.
Sites
10
5
20
27
81
154
351
156
105
Parameter Median BOD (mg O2/l)
Median 4.00 4.50 2.70 2.70 2.10 1.70 1.60 1.50 1.30
Mean 8.35 4.28 3.04 2.87 2.31 1.95 1.67 1.49 1.41
SD 10.23 1.36 1.60 1.41 1.05 0.92 0.42 0.45 0.37
Parameter Median Unfiltered Molybdate Reactive Phosphorus (mg P/l)
Median 0.209 0.153 0.135 0.130 0.070 0.043 0.030 0.020 0.015
Mean 0.681 0.159 0.184 0.187 0.116 0.063 0.047 0.027 0.022
SD 1.240 0.033 0.174 0.195 0.145 0.076 0.054 0.024 0.019
Parameter Median Oxidised Nitrogen (mg N/l)
Median 1.85 1.19 1.67 1.79 1.70 1.50 1.20 0.88 0.54
Mean 2.33 1.44 1.68 2.36 2.02 1.62 1.55 1.29 0.76
SD 1.62 0.83 0.91 2.37 1.53 1.21 1.31 1.14 0.77
Parameter Median Ammonia (total) mg N/l
Median 0.380 0.380 0.220 0.190 0.080 0.050 0.040 0.030 0.030
Mean 1.012 0.307 0.376 0.385 0.152 0.082 0.062 0.046 0.049
SD 1.220 0.197 0.538 0.497 0.265 0.104 0.077 0.048 0.073
The National Biological Survey of Rivers, in progress since 1971, has
provided a unique time series and a useful set of historical conditions and
trends in Irish rivers. In many cases the historical survey results can be used
to provide reference conditions for the purposes of the EU’s Water
Framework Directive.
7
The EPA’s river biologists have provided a range of high status and
impacted sites for this project based on their experience of Irish rivers.
There is a need for an in-depth analysis of the impacts of eutrophication on
river ecosystems and in particular, there is a need to be able to quantify the
changes as a river ecosystem departs from its high status or reference
condition as it is termed in the EU Water Framework Directive. The sites
selected and described below form a basis for undertaking such an analysis
and this forms a major part of the current study.
Serious organic pollution has declined steadily in Irish rivers since the
1970s and now accounts for approximately one percent of polluted river
channel. In parallel with this, however, a second trend of increasing slight
and moderate levels of pollution has occurred resulting in some 30% of
surveyed river channel now being classed as slight or moderately polluted
(Clabby et al., 1992; McGarrigle et al., 2002).
Eutrophication is regarded as the main reason for this decline in water
quality in Irish rivers with a notable loss of species and a widespread change
in trophic status over a 30-year period (McGarrigle, 2001). It is vital,
therefore, in establishing reference conditions to understand the detailed
mechanisms by which eutrophication affects river ecology. Thus, in setting
reference conditions for rivers, it is important to have a more precise
understanding of the driving forces behind the long-term changes that have
occurred already in Irish rivers and in macroinvertebrate communities
particularly as they are the most important element defining Q-values in the
Irish Quality Rating System. The high status sites chosen for this project
had ideal conditions for the most sensitive taxa. Some of the eutrophic sites
contained these sensitive indicators but only sporadically during the study
programme while other eutrophic sites did not have any of these taxa at all.
National statistics and long term trends are largely based on the EPA Q-
values. The Q-values in turn are based on the response of
macroinvertebrates especially but also of phytobenthos and macrophytes in
rivers to pollutants such as phosphorus, and inputs of organic matter or toxic
8
substances such as ammonia, heavy metals and pesticides. Similarly, they
respond to physical factors such as substratum, flow velocity, siltation
channel alteration, erosion and temperature effects. Finally, the influence of
biotic factors such as predation and inter-specific competition will also play
a role in defining the final ecological community of inhabiting a river
stretch. Because the Water Framework Directive requires the definition of
reference conditions for different ecological types and regions, it is
particularly important to be able to understand and define the reasons for
departure from reference conditions causing loss of sensitive indicator
species such as Ecdyonurus.
9
1.4 The Water Framework Directive
1.4.1 Background
As part of a substantial restructuring of EU water policy and legislation, a
Directive establishing a new framework for community action in the field of
water policy (2000/60/EC; European Commission 2000) was agreed by the
European Parliament and Council in September 2000 and came into force
on 22nd
December 2000. The Directive, generally known as the Water
Framework Directive (WFD), rationalises and updates existing water
legislations and provides for water management on the basis of River Basin
Districts (RBD’s). It outlines a legal structure for the assessment of all
types of water bodies in Europe.
The purpose of this Directive is to establish a Framework for the protection
of inland surface waters, transitional waters, coastal and groundwaters. A
focus of the assessment systems demanded by the WFD is the use of biotic
indicators (macrobenthic fauna, fish fauna and aquatic flora). It is
potentially the most significant piece of legislation ever to be enacted in the
interests of conservation of freshwater and saline ecosystems. It seeks to
incorporate existing legislation into the new framework with bioassessment
as the key approach. The Directive uses biological measures to a greater
extent than previous directives such as the Freshwater Fish Directive, where
water quality was based predominately on chemical determinands. There is
quite a distinction between water quality and ecological status and systems
used currently to establish the former are only a small part of those needed
for the latter (Moss et al., 2001).
1.4.2 Ecological boundaries
The Directive requires catchments to be managed in a holistic manner
whose ultimate effectiveness will be reflected in the degree to which aquatic
habitats are restored to ‘good’ ecological status (Moss et al., 2001).
Maintaining ‘high’ ecological status, where it already exists, or achieving
‘good ecological status’ is the ultimate goal. One of the first requirements
10
of the Directive is the development of a typology among flowing, standing,
transitional and coastal waters. Within each of these groups, ecotypes must
be defined and the pristine (reference) characteristics of each type defined.
On establishing individual ecotypes, each is classified into a system
according to deviations from this reference condition (high quality).
Furthermore, the ecological status of a water body is defined by comparing
the biological community composition present with the near-natural
reference condition (Hering et al., 2004). Reference conditions for Irish
rivers will be included in the forthcoming characterisation report for Ireland
to be completed under Article 5 of the WFD by 22nd
March 2005.
The classification system includes ‘good’, ‘moderate’, ‘poor’, and ‘bad’
quality and must be established by comparing biological elements with their
reference condition and taking into account supporting chemical and
morphological characteristics. As mentioned above, WFD includes the
concept of “no deterioration” in ecological quality but the main aim is to
generate plans for the gradual improvement of the ecological quality of all
surface waters until they achieve “good” status. Within the WFD, the aim
to provide a more sustainable water system is based on the view that good
ecological quality is more “natural” and therefore more sustainable (Logan,
2001). Methods for WFD compliant classification of ecological status of
each of the biotic elements listed in Annex V of the WFD are currently
under development in Ireland and elsewhere across Europe in preparation
for the Directive’s monitoring programmes, which must commence in 2006.
1.4.3 The Reference condition
A central part of the EU WFD is the identification of expected background
reference conditions with no or minimal anthropogenic stress. According to
the Directive, Member states are required to identify reference conditions
for the purpose of defining a reference biological community (European
Commission, 2000; REFCOND, 2003). There have been a number of
definitions of “reference” conditions put forward to date:
11
• Expected background (i.e. reference) conditions with no or minimal
anthropogenic stress and satisfying the following criteria: (1) they
should reflect totally, or nearly, undisturbed conditions for
hydromorphological elements, general physicochemical elements, and
biological quality elements, (2) concentrations of synthetic pollutants
should be close to zero or below the limit of detection of the most
advanced analytical techniques in general use, and (3) concentrations of
specific non-synthetic pollutants, should remain within the range
normally associated with background levels (European Commission,
2000; REFCOND, 2003)
• Representing important aspects of ‘natural’ or pre-Columbian conditions
and at the same time, politically palatable and reasonable (Hughes,
1995)
• The condition that is representative of a group of minimally disturbed
sites organised by selected physical, chemical and biological
characteristics, (Reynoldson et al., 1997)
• Regionally-representative sites that are indicative of expected
background conditions in the absence of anthropogenic stress (Johnson,
1999)
The idea of a reference condition is a critical element in approaches now
being developed for biomonitoring and bioassessment of aquatic system. In
America for instance, the reference condition is central to currently accepted
ideas of ‘biocriteria’ being developed by the US Environmental Protection
Agency (EPA) (Davis and Simon, 1995). A similar approach has been used
in the UK for river classification and water-quality assessment (Wright,
1995). It is also being used in Canada to develop sediment guidelines for
the Great Lakes (Reynoldson et al., 1995) and is the basis for the National
River Health Program in Australia (Parsons and Norris, 1996). Ideally, a
river is assessed and its ecological status is compared with the reference
state (pristine state). The assumption being that the reference state is in
itself sustainable because anthropogenic impacts are “minimal” or as maybe
the case, they are simply more sustainable than an impacted situation
(Logan, 2001).
12
1.4.4 Classifying ecological status
Due to anthropogenic impacts on the environment it can often be quite
difficult to find pristine rivers of high ecological status. The definition of
ecological status itself relates to the observed status of the site when
compared with a reference status. In describing reference conditions, water-
quality monitoring is required based on a pre-established criteria that exist at
a wide range of sites rather than relying on information from one or a few
control sites. These reference conditions are based on spatial, temporal or
ecologically modelled conditions where human impacts are minimal and the
reference conditions then serve as the control against which test-site
conditions are compared. The idea of the reference condition is really one
of the best available condition and it is represented by information from
numerous test sites (Reynoldson et al., 1997). Many countries, however,
would like to use the best available condition but in many cases the best
available may actually be quite poor. Historical data, expert opinion,
models, palaeolimnology and outside-state reference sites may be used to
recreate reference conditions rather than just accept the best available.
Establishing baseline reference conditions against which to measure the
effects of human activities is only the first step in the WFD’s requirement of
being able to determine the ecological status of inland waters. The WFD
also stipulates that the level of human impact on the structure and function
of aquatic ecosystems needs to be defined in terms of ecological quality
elements (Johnson, 2001). There are five biological quality elements in
Annex V of the WFD that require examination: macroinvertebrates,
phytobenthos, phytoplankton, macrophytes and fish fauna. In addition to
this, member states must be able to identify five levels of impact (or
ecological status) using these ecological quality elements.
In Irish terms, the ecological status of a river is classified into the following
categories based on the Q-value Rating System:
13
• Q5, Q4-5 classified as ‘High’ status
• Q4 classified as ‘good’ status
• Q3-4 classified as ‘moderate’ status
• Q3, Q2-3 classified as ‘poor’ status
• Q1, Q1-2, Q2 classified as ‘bad’ status
The final classification will, however, depend on the results of the official
EU intercalibration exercises. In addition to measuring quality elements or
organism groups for classifying the ecological status, the WFD also
specifies that a number of hydromorphological and physico-chemical
elements need to be examined. This in turn will support and interpret the
organism-based monitoring/assessment decision of whether or not a site
deviates from good ecological quality. Good ecological status does not
differ significantly from undisturbed, high ecological status or reference
conditions, whereas moderate, poor and bad ecological quality differ from
the expected and the biological communities show varying degrees of
human induced stress ranging from moderate to severe changes in
community structure and composition (Johnson, 2001).
1.4.5 Benthic macroinvertebrates
Benthic macroinvertebrates, together with algae, serve as the most common
groups used for assessing the ecological river quality (Hellawell, 1986; De
Pauw and Hawkes, 1993; Rosenberg and Resh, 1993). Generally, benthic
macroinvertebrates are capable of reflecting different anthropogenic
perturbations through changes in structure or function in the assemblages
and thus enable an overall assessment of ecological status. Besides organic
pollution, which can be assessed using a large number of biological indices,
benthic macroinvertebrates can also be used to detect acid stress, habitat
loss and overall stream degradation (Hering et al., 2004). For these reasons,
benthic macroinvertebrates are likely to play a major role in future stream
assessment in coherence with the Water Framework Directive.
The individual European countries have very different traditions in the use
of benthic macroinvertebrates for biomonitoring purposes. Overviews were
14
given by Knoben et al., (1995), Nixon et al., (1996) and Birk and Hering
(2002). The differences in the methods are inherent in the intensity of
sampling and sorting, the taxonomic resolution, the storing and the
statistical treatment of the data, the metrics used for assessment and the
quality assurance procedures used in this process. Therefore, comparisons
of results obtained with different national methods are complex and in many
cases hindered by these differences. While existing methods in some of the
countries are partly fulfilling the requirements of the Water Framework
Directive (UK: Wright et al., 2000; France: Agences de l’Eau, 2000),
existing methods need adaptation in other countries (Germany: DEV, 1992;
Austria: Austrian Standards M 6232, 1997; Chovanec et al., 2000;
Netherlands: Peeters et al., 1994; Italy: Ghetti, 1997) or no official method
is available (Greece, Portugal). In Ireland’s case the Quality Rating System
must be made WFD compliant by converting it to an ecological quality ratio
(EQR) system with a score that ranges from 0 to 1.
1.4.6 Conclusions
The WFD is a wide ranging and ambitious piece of European environmental
legislation setting clear objectives to ensure that ‘good status’ is achieved
for all European Waters by 2015. The aims of the WFD present major
challenges to everyone involved in protection, use and management of the
aquatic environment. Significant resources have been mobilised to meet the
challenges and meet the tight time schedule for the activities, which must be
undertaken. It is hoped that the current study can help in the overall
development of a WFD-compliant macroinvertebrate monitoring system by
helping to better understand the response of some of the key indicator
species to eutrophication and other pressures.
15
1.5 Eutrophication
1.5.1 Background
Eutrophication, whether in a river or lake, is the enrichment of water by
plant nutrients that are normally in short supply, which consequently cause
an imbalance in the food web, resulting in high levels of plant and algal
growth. The predominant nutrients that control plant and algal growth, are
phosphorus and nitrogen. Phosphorus is usually available to aquatic
organisms in water as phosphates and since it is naturally present in lower
concentrations than nitrogen, phosphate usually acts as the limiting growth
factor. In Ireland, one of the most serious environmental pollution problems
is the over-enrichment of surface waters by phosphorus and to a lesser
extent, nitrates (Environmental Protection Agency, 2002).
1.5.2 Nutrient sources
This type of pollution poses a threat to Ireland’s game fish populations and
has resulted from excessive inputs of nutrients from a number of sources.
Estimates indicate that agriculture is responsible for the largest inputs of
phosphorus and nitrates to waters. Sewage is another major source of these
nutrient inputs. These sources are usually classified according to the type
of discharge, namely ‘point’ and ‘non-point’ sources. Point sources include
discharges from sewage and the run-off from open farmyards. Non-point
sources originate by diffuse losses from land to which excessive amounts of
animal waste or artificial fertilisers rich in organic phosphates have been
applied. Phosphates are also derived from natural sources due to erosion.
This source is generally not as biologically available for plant growth and is
usually a very small source in comparison with anthropogenic sources.
Rainfall and dry deposition from the atmosphere are further sources but
generally to a lesser extent (Lucey et al., 1999).
Large point sources such as sewage discharge are often the major, and more
easily controlled, source of phosphorus loads but diffuse agricultural
sources can be even larger than municipal sources especially in less densely
16
populated countries such as Ireland. There is now a significant body of
evidence to suggest that phosphorus from agricultural sources can represent
a significant input to freshwater and the increase in phosphorus
concentrations in agricultural drainage water over time reflects the
accumulation of phosphorus in soils (Sharpley et al., 2000). Even losses
from Irish grasslands, without any farmyards, can reach 3kg P/ha/annum
where the soil burden is high and the soil type is prone to losses (Jordan et
al., in press). This is many times the typical background or reference
condition loss rates of less than 0.1 kg P/ha/annum.
1.5.3 Eutrophication Trends in Ireland
National river quality surveys were first carried out in Ireland in 1971 on a
river channel length of 2,900km. It became apparent in the late 1970s that
river eutrophication was an actual or potential problem in many river
catchments (McGarrigle, 2001). One of the first studies to highlight this
problem was that undertaken for the original water quality management
plans (WQMP), particularly those prepared for the south-eastern rivers like
the River Suir (e.g. Horkan, 1980). By the early 1980s, there was a
widespread national trend of increased eutrophication in rivers in Ireland.
This prompted a series of studies aimed at gaining a better understanding of
river eutrophication which would help to control and reverse the increasing
trend of eutrophication in Irish rivers (e.g. McGarrigle 1983, 1984a, 1984b;
McGarrigle and Lucey 1989, 1990; McGarrigle et al., 1987).
The Environmental Protection Agency expanded its river quality
investigations to 12,700km between 1991-1994 and to 13,500km between
the period 1995 to 1997 (Lucey et al., 1999). The third biological
investigation was carried out during the period 1998-2000 in which the
national baseline of some 13,100km of river channel was assessed. The
most recent assessment of river quality in Ireland shows an improvement in
water quality for the first time since the surveys began (Environmental
Protection Agency, 2002). Unpolluted channel length increased from 67 %
in the period 1995-1997 to almost 70% in the period 1998-2000. However,
the overall status is still unacceptably poor, in comparison to what it was 20
years ago.
17
Over 30% of the national river channel is now considered to be polluted to
some extent. This pollution is attributed in the main to eutrophication. The
degree of pollution is minor in many cases but it is of concern in view of its
potential impact on the pollution-sensitive trout and salmon in rivers and
lakes. In Ireland this is of particular concern because most rivers are
salmonid waters, capable of supporting game fish such as trout, salmon or
sea trout. Luxuriant plant growth in salmonid rivers may cause diurnal
fluctuations in Dissolved Oxygen (DO) resulting in fish kills due to low DO
levels, particularly the pre-dawn period. In 1990, Moriarty found that up to
50% of fish-kills reported in Ireland were due to this type of phenomena in
eutrophic waters during warm weather.
It is worth noting that the observed improvements have mainly occurred in
catchments that have had intensive management programmes implemented
over the last five to seven years (Environmental Protection Agency, 2002).
The recent EU Water Framework Directive provides for a more integrated
approach to controlling water pollution. Its full implementation will be a
major policy challenge for Ireland, but should ultimately lead to a
significantly improved water quality and water management across the
State.
1.5.4 Effect of eutrophication on macroinvertebrates
As with many other environmental stresses two interrelated trends are
readily discerned when increasing enrichment occurs. The first is a
reduction in the diverse macroinvertebrate community characteristics of
clean water, in which many species are represented by relatively few
individuals, towards a condition, in which under the influence of severe
pollution, a few species are represented by very large numbers of
individuals. These species are those which are able to take advantage of the
changes which the pollutant induces and to exploit the increased food
supply which is provided. The second change, which is more relevant to
that which occurs during eutrophic conditions, is the progressive
disappearance of particular indicator species until very few remain and their
place is taken by species not previously present or, at least, not abundantly
18
present (Hellawell, 1986). In Ireland, Ecdyonurus is a key water quality
indicator genus and little is known about the factors responsible for its
disappearance from eutrophic or other polluted rivers.
Although intermediate levels of organic enrichment may favour certain
suspension or deposit feeding macroinvertebrate groups (such as blackfly
and chironomid larvae), changes in the substratum (through increased
sedimentation of organic matter) and low dissolved oxygen concentrations
that often occur at high levels of organic pollution and under eutrophic
conditions, usually result in the disappearance of intolerant taxa (Hynes,
1960; Hellawell, 1986). It is worth noting that during the preliminary stages
of eutrophication, species numbers may increase due to favourable
conditions for the particular invertebrate species.
While the Q-value System was originally based on the sensitivity of
invertebrates to organic pollution, it has been found that the invertebrates
show a definite response to eutrophication, which is related to but separate
from, organic pollution. Rivers with low BOD and other organic pollutants
like ammonia may suffer from diurnal oxygen fluctuations due to nutrient-
induced growth of excessive plant biomass resulting in loss of sensitive
species. As eutrophication progresses, other symptoms may arise, for
instance causing changes in algal species or resulting in habitat siltation
(McGarrigle, 2001).
1.5.5 Diurnal dissolved oxygen (DO) variation
One of the main impacts of eutrophication is the increase in primary
production, combined with respiration, giving rise to a diurnal pattern of
dissolved oxygen concentrations in rivers (Odum, 1956; Edwards and
Owens, 1965). Dissolved oxygen depletion usually occurs at night in these
rivers caused by the respiratory demands of the plant biomass and sediment
bacteria. A clean river will normally display a DO reading near the 100%
saturation level with little deviation above or below this level over the day
in response to photosynthesis and respiration of plant material (McGarrigle,
1998).
19
A river unaffected by eutrophication will have low to medium plant biomass
and the phosphorus levels will be low. Since phosphorus is normally the
limiting nutrient, the addition of bioavailable phosphorus will cause an
increase in plant growth. It has been shown that the variations in DO levels
increases as the plant biomass increases causing diurnal fluctuations thereby
exerting a significant strain on salmonid populations. This may be
somewhat overlooked in routine monitoring programmes as practically all
national and local river monitoring programmes over the past 30 years,
measure day time DO levels only (McGarrigle, 2001).
20
1.6 Research Objectives
The status of Ecdyonurus as a key indicator species is based largely on
empirical evidence. In particular there are clear statistical links between the
presence and absence of Ecdyonurus and water quality parameters such as
BOD, ammonia, nitrate and phosphate that are important chemical
indicators of water quality (McGarrigle, 1998). The Irish Phosphorus
Regulations are based on such a link between biological Q-values and
phosphate concentrations. While these strong empirical links exist, the
precise mechanisms controlling the distribution of Ecdyonurus in eutrophic
and polluted rivers are not that well understood.
Ecdyonurus is one of the most useful indicators of pollution because it is
almost ubiquitous in unpolluted waters, but is sensitive to pollution. It can
be expected to occur in river types ranging from large to small, from hard
water to moderately acidic ones and from fast flowing to all but the slowest-
flowing stretches of river, provided the river is not polluted.
In the light of the Water Framework Directive’s need for ecological
assessment of rivers, it will be necessary to adapt the Irish Q-value System
in order to provide a broader ecological assessment of river status. A more
comprehensive ecological assessment needs to take into account, for
example, the expected faunal and floral composition for a given type of site
under pristine conditions and factors such as river hydromorphology and
invasive species.
An important objective of this project is to obtain an improved
understanding of reference conditions as revealed by changes in the
macroinvertebrate fauna and phytobenthos when a river’s status begins to
depart from its pristine state. To this end, this project attempts to refine our
knowledge of the ecology of Ecdyonurus and other members of the
Heptageniidae family in relation to a range of potential controlling factors in
the riverine environment.
21
1.7 Study locations
A GIS map outlining the ten study locations and catchment geology selected
for this study is shown in Fig. 1.1. The five high status and five impacted
rivers and associated information are shown in Table 1.2 and Table 1.3,
respectively.
Each of the ten rivers are described separately in detail below followed by
the catchment characteristics showing CORINE 2000 landuse, GSI rock
units (bedrock geology) and river network indicating Strahler stream order
(Fig. 1.2 -Fig. 1.11).
22
Table 1.2 List of the five high status rivers studied in this investigation.
River Code Station
number
Location GPS Q-value
2001-2003 Owengarve 34O03 0200 Bridge upstream of
the Moy River
confluence (Dawros
Bridge)
54 00 784 N
8 50 154 W
5
Castlebar 34C01 0020 Bridge near Graffa
More upstream of
Lough Mallard
53 51 42 N
9 21 00 W
4-5
Brusna (North Mayo) 34B07 0400 Bridge west of
Cloonta
54 07 58 N
9 04 10 W
4-5
Dunneill 35D06 0100 Bridge 2km
upstream Dromore
West
54 14 15 N
8 51 42 W
5
Callow Loughs Stream 34C08 0300 Bridge upstream
Yellow River
54 01 20 N
9 02 00 W
4-5
Table 1.3 List of the five impacted rivers studied in this investigation.
River Code Station
number
Location GPS Q-value
2001-2003 Cartron 33C02
0100 Bridge west of
Lough Gall
53 56 13 N
9 49 38 W
3-4
Lough na Corralea Stream 30L03
0400 Bridge east of
Cloonee
53 42 51N
9 19 02 W
3
Robe 30R01 0400 Hollymount Bridge
53 39 42 N
9 07 19 W
3
Mullaghanoe
34M03 0100 Bridge WNW of
Bellahy
53 58 04 N
8 48 01 W
3
Mad
34M04 0100 Bridge in
Cloonacool
54 06 07 N
8 46 23 W
3
23
Fig. 1.1 GIS map showing the location of the ten river catchments and associated geology.
Cartron River
Robe River
Castlebar River
Owengarve River
Dunneill River
Lough na Corralea Stream
Brusna River
Mullaghanoe River
Mad River
Callow Loughs Stream
24
1.7.1 Site 1: The Owengarve River
The Owengarve River site is situated around 5km west of Curry village in
Sligo. The study site (EPA national river code 34O03-0200), was located
approximately 100m downstream of the bridge which is situated upstream
of the Moy River confluence. Monthly samples were taken in an area
measuring approximately 15m x 6m (depending on the water level on the
day of sampling). Its co-ordinates are E: 145303, N: 3074170; Latitude: 54°
00’ 784” N, Longitude: 8° 50’ 154” W. The Owengarve is a fourth order
river with a catchment area of 121.27 km2. The catchment is dominated by
pasture and peat bog.
The average wet width was 6-12m for the duration of the sampling period.
Pebbles and cobble were the main substratum type in this river. The
dominant tree species present were alder (Alnus spp.), willow (Salix spp.),
Birth (Betula spp.) and hawthorn (Crataegus monogyna). The vegetation
along the banks consisted mainly of mint (Mentha spp.), butterbur (Petasites
hybridus), puccinellia (Glyceria spp.) and forget-me-not (Myosotis spp.).
The main species of emergent macrophytes present were bur-reed
(Spraganium erectum), duckweed (Lemna spp.), watercress (Nasturtium
spp.), wild celery (Apium spp.) and horsetail (Equisetum spp.). The floating
plants found were pondweed (Potamogeton natans) and iris (Iris
pseudacorus). The main submerging plants were water-milfoil
(Myriophyllum spp.), Fontinalis spp. and other moss varieties.
Monitoring by the EPA began in the Owengarve River in 1980 and it
continues to retain a Q-value of 5.
Fig. 1.2 outlines the catchment characteristics of the Owengarve River
showing the CORINE 2000 landuse, GSI rock units (bedrock geology) and
the river network indicating Strahler stream order.
25
0 6 12 Kilometers
N
EW
S
0 6 12 Kilometers
N
EW
S
0 6 12 Kilometers
N
EW
S
Fig. 1.2 Catchment characteristics for Catchment of the Owengarve River Sample Site 34O030200
showing CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler
stream order.
Sampling location
26
1.7.2 Site 2: The Castlebar River
The Castlebar River site is located approximately 6km west of Castlebar
town. The study site (EPA national river code 34C01-0020), was located
approximately 200m downstream of the bridge near Graffa More, upstream
of Lough Mallard. Monthly sampling was carried out in an area measuring
approximately 9m x 3m (depending on the water levels on the day of
sampling). Its co-ordinates are E: 110930, N: 291732; Latitude: 53° 51’ 20”
N, Longitude: 9° 20’ 55” W. The Castlebar River is a second order river
with a catchment area of 5.25 km2. Its catchment is dominated by
agricultural land and peat bog with a small forest located a few miles
upstream of the sampling site.
The average wet width of the river during the sampling period was 3-4m.
Pebbles and cobbles dominated the riverbed substrate. The main riparian
trees present were hawthorn, willow and ash (Fraxinus excelsior). The
vegetation on the riverbanks consisted of montbretia (Crocosmia x
crocosmiflora) while bur-reed was the main emergent macrophyte. The
submerged macrophytes were Fontinalis spp and other mosses.
The quality of this river is good and has maintained a Q-value of 4-5 since
monitoring began in the 1970s.
Fig. 1.3 outlines the catchment characteristics of the Castlebar River
showing the CORINE 2000 landuse, GSI rock units (bedrock geology) and
the river network indicating the Strahler stream order.
27
#
#
2 0 2 4 Kilometers
0 2 4 Kilometers
N
EW
S
0 2 4 Kilometers
N
EW
S
Fig. 1.3 Catchment characteristics for Catchment of the Castlebar River Sample Site 34C010020 showing
CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler stream
order.
Sampling location
28
1.7.3 Site 3: The Brusna River
The Brusna River is situated approximately 7km north east of Ballina town,
in north Mayo. The study site (EPA national river code 34B07-0400), was
located 10-20m downstream of the bridge west of Cloonta. Monthly
sampling was carried out in an area measuring approximately 17m x 6m
(depending on the water level on the sampling day). Its co-ordinates are
E: 130196; N: 320988; Latitude: 54° 07’ 58” N, Longitude: 9° 04’ 10’’ W.
The Brusna is a fourth order stream with a catchment area of 28.15 km2.
Pasture and peat bogs are the main land use.
The average wet width of the river was 6-8m during the study period. The
substratum consisted mainly of large and small cobble, pebbles and sand
and gravel. The dominant tree species were sycamore (Acer
pseudoplatanus), willow, ash and alder. The vegetation along the banks of
the river consisted of meadow sweet (Filipendula ulmaria), willow-herb
(Epilobium spp), purple loosestrife (Lythrum salicaria), snowberry
(Symphoricarpos albus), Petasites spp and montbretia. The emergent plants
in this river were bur-reed, wild celery, mint and watercress. The main
submerged macrophytes were Fontinalis spp, other species of moss and
water-milfoil. The floating Pondweed was also present.
The Brusna River has maintained good water quality standards since
monitoring commenced in the late 1980s and presently holds a Q-value of
4-5.
Fig. 1.4 outlines the catchment characteristics of the Brusna River showing
the CORINE 2000 landuse, GSI rock units (bedrock geology) and the river
network indicating Strahler stream order
29
0 3 6 Kilometers
N
EW
S
0 3 6 Kilometers
N
EW
S
#
#
#
#
#
#
#
7 0 7 Kilometers
Fig. 1.4 Catchment characteristics for Catchment of the Brusna River Sample Site 34B070400 showing
CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler stream
order.
Sampling location
30
1.7.4 Site 4: The Dunneill River
The Dunneill River is located approximately 2km south, south east of the
village of Dromore West in Sligo. The study site (EPA national river code
35D06-0100), was located roughly 40m downstream of the bridge at
Dromore West. Monthly samples were taken in an area measuring
approximately 14m x 8m (depending on the water level on the day of
sampling). Its co-ordinates are E: 144492; N: 329228; Latitude: 54° 14’
15’’ N, Longitude 8° 51’ 42’’ W. The Dunneill is a third order stream with
a catchment area of 12.02 km2. Pasture and peat bogs are the dominant land
uses in the catchment.
During the sampling period, the average wet width of the river was
approximately 8-10m. The dominant substrate in the river consisted of
pebbles and cobbles and to lesser extent boulders. The mains tree species
were ash, alder, willow and fuchsia (Fuchsia magellanica). The bankside
vegetation consisted of rush (Juncus spp.) and Petasites spp. The main
emergent plants were mint, Carex spp. and Caltha spp., while moss was the
dominant submerged macrophyte. The land use in this catchment is mainly
pasture.
The Dunneill River has maintained a high water quality standard since the
monitoring programme employed by the EPA commenced. It continues to
have a Q-value of 5.
Fig. 1.5 outlines the catchment characteristics of the Dunneil River showing
the CORINE 2000 landuse, GSI rock units (bedrock geology) and the river
network indicating Strahler stream order
31
0 3 6 Kilometers
N
EW
S
0 3 6 Kilometers
N
EW
S
0 3 6 Kilometers
N
EW
S
Fig. 1.5 Catchment characteristics for Catchment of the Dunneill River Sample Site 35D060100 showing
CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler stream
order.
Sampling location
32
1.7.5 Site 5: Callow Loughs Stream
Callow Loughs Stream is located approximately 3 km north east of Foxford
town in Mayo. The study site (EPA national river code 34C08-0300), was
located just downstream of the bridge situated upstream of the Yellow River
confluence. Monthly sampling was carried out in an area approximately
measuring 7m x 15m (depending on the water level on the sampling day).
Its co-ordinates are E: 129356, N: 305629; Latitude: 54° 01’ 120” N,
Longitude: 9° 02’ 00” W. Callow Loughs Stream is a second order stream
with a catchment area of 8.76 km2. The predominant land use in the area is
pasture and there is also a working quarry upstream and a disused one
downstream.
The average wet width of the river was between 6-8m for the sampling
duration. Large and small cobbles were the main substratum types. The
species of trees consisted of a mixture of alder, ash and willow. Bankside
vegetation consisted mainly of hart’s tongue (Asplenium spp.), blackthorn
(Prunus spinosa) and fern species. The emergent macrophytes were mint,
wild celery and speedwell (Veronica spp.). Moss and water-milfoil were the
most common submerged macrophyte species.
Callow Loughs Stream has good water quality and since monitoring began
in 1989, it fluctuated between Q4 and Q4-5. Since 1998, it continued to
maintain a Q4-5 rating.
Fig. 1.6 outlines the catchment characteristics of Callow Loughs Stream
showing the CORINE 2000 landuse, GSI rock units (bedrock geology) and
the river network indicating Strahler stream order
33
#
#
2 0 2 4 Kilometers
0 2 4 Kilometers
N
EW
S
0 2 4 Kilometers
N
EW
S
Fig. 1.6 Catchment characteristics for Catchment of the Callow Loughs Stream Sample Site 34C080300
showing CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler
stream order.
Sampling location
34
1.7.6 Site 6: The Cartron River
The Cartron River is located approximately 4km north west of Mulranny
village in Co.Mayo. The study site (EPA national river code 33C02-0100)
was located roughly 100m upstream of the bridge west of Lough Gall.
Monthly sampling was carried out in a designated area measuring
approximately 15m x 8m (depending on the water level on the day of
sampling). Its co-ordinates are E: 80000, N: 300184; Latitude: 53° 56’
13’’N, Longitude: 9° 49’ 38’’ W. The Cartron River is a fourth order river
with a catchment area of 10.47 km2. A large forestry plantation upstream
and peat bog surrounding the river dominate the land use.
The average wet width of the river during the sampling programme was 6-
12m. The substratum of the Cartron River was dominated by large and
small cobble interspersed with gravel. A large conifer plantation was
located upstream of the river site. The bankside vegetation consisted of
rush. There were no emergent macrophytes and moss and liverworths were
the only submerged vegetation in this river.
In 1982, the Cartron was assigned a Q-value of 4-5 and retained this
standard until it dropped to a Q4 when assessed in 1994. The quality
declined thereafter to a Q3-4 in 1997 and from 1999 to the present study, it
was assigned a Q3. There has been a notable decline in the numbers of
ephemeropteran species over the years. This river is prone to quite a
substantial amount of algal growth particularly during the spring and
autumn months.
Fig. 1.7 outlines the catchment characteristics of the Cartron River showing
the CORINE 2000 landuse, GSI rock units (bedrock geology) and the river
network indicating Strahler stream order
35
0 2 4 Kilometers
N
EW
S
0 2 4 Kilometers
N
EW
S
0 2 4 Kilometers
N
EW
S
Fig. 1.7 Catchment characteristics for Catchment of the Cartron River Sample Site 33C020100 showing
CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler stream
order.
Sampling location
36
1.7.7 Site 7: Lough na Corralea Stream
Lough na Corralea Stream is situated approximately 2km north west of
Party in Co. Mayo. The study site (EPA national river code 30L03-0400)
was located just downstream of the bridge east at Cloonee. Monthly
sampling was carried out in an area measuring approximately 16m x 4m
(depending on the water level on the day of sampling). Its co-ordinates are
E: 113055; N: 274654, Latitude: 53° 42’ 51’’N, Longitude: 9° 19’ 02’’ W.
Lough na Corralea Stream is a fourth order river with a catchment area of
12.58 km2. Pasture is the dominant land use.
The average wet width in the river during the study programme was
between 5 and 7m. Lough na Corralea Stream is a fast flowing stony river
system with quite a lot of bedrock and large boulders but is largely
dominated by cobble and gravel. The main riparian trees are sycamore,
willow and ash. The main vegetation growing on the riverbanks were
montbretia and mint. The submerged macrophytes consisted of rush, moss
and Callitriche spp. The emergent plants in this river were wild celery,
pondweed and water-plantain (Alisma spp.).
The quality of Lough na Corralea Stream has varied from a Q4 in 1989 to a
Q4-5 in 1996. In 2000, the river was assessed and assigned a Q3-4 with a
distinct absence of species from the Heptageniidae family. (An Oscillitoria
type of algal mat grows readily on the substrate in this river).
Fig. 1.8 outlines the catchment characteristics of Lough na Corralea stream
showing the CORINE 2000 landuse, GSI rock units (bedrock geology) and
the river network indicating Strahler stream order
37
0 2 4 Kilometers
N
EW
S
0 2 4 K i l o m e t e r s
#
4 0 4 Kilometers
Fig. 1.8 Catchment characteristics for Catchment of Lough na Corralea Stream Sample Site 30L030400
showing CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler
stream order.
Sampling location
38
1.7.8 Site 8: The Robe River
The Robe River flows through the village of Hollymount, which is located
approximately 6km north east of the town of Ballinrobe in Co. Mayo. The
study site (EPA national river code 30R01-0400) was located approximately
12-15m downstream of the bridge at Hollymount. Its co-ordinates are
E: 125839, N: 268640; Latitude: 53° 39’ 42’’N, Longitude: 9° 07’ 19’’W.
Monthly sampling was carried out in an area measuring approximately 12m
x 25m (depending on the water level on the day of sampling). The Robe
River is a fourth order river with a catchment area of 193.20 km2. The
predominant land use is pasture where sheep and cattle graze close by.
The average wet width in the river during the study period ranged between
12-15m. Cobble and gravel dominated the substrate with boulders
interspersed. There was a high content of silt and gravel in this river,
particularly around the marginal areas. The riparian vegetation consisted of
gorse (Ulex europaeus), alder and hawthorn. The submerged macrophytes
present were forget-me-not and water-milfoil. The main emergent plants
were watercress, bur-reed and Scirpus.
In 1984, weirs were constructed just downstream of the bridge where the
study site is located which appear to have changed the energy of the river
and in spite of the presence of good riffled areas, the water quality of this
river has deteriorated over the last few decades from a Q4 to a Q3 with a
substantial loss of sensitive indicator species. This river is prone to
eutrophication quite regularly.
Fig. 1.9 outlines the catchment characteristics of the Robe River showing
the CORINE 2000 landuse, GSI rock units (bedrock geology) and the river
network indicating Strahler stream order
39
0 8 16 Kilometers
N
EW
S
0 8 16 Kilometers
N
EW
S
0 8 16 Kilometers
N
EW
S
Fig. 1.9 Catchment characteristics for Catchment of the Robe River Sample Site 30R010400 showing
CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler stream
order.
Sampling location
40
1.7.9 Site 9: The Mullaghanoe River
The Mullaghanoe River is situated 0.5km west north west of Bellahy village
in Co. Mayo. The study site (EPA national river code 34M03-0100) was
located approximately 2m downstream of the bridge. Monthly sampling
was carried out in a designated area measuring approximately 10m x 4m
(depending on the water level on the day of sampling). Its co-ordinates are
E: 147505, N: 302542, Latitude: 53° 58’ 04’’N, Longitude: 8° 48’ 01’’ W.
The Mullaghanoe River is a fourth order river with a catchment area of
25.19 km2. Pasture and peat bogs are the main land use within the
catchment.
The average wet width during the sampling programme was 4-8m. Large
and small cobble and gravel dominated the substratum type. The main tree
species were ash, willow, sycamore and hawthorn. The bankside vegetation
consisted of Glyceria, willow herb (Epilobium), meadowsweet (Airgead
luachra) and purple loosestrife (Lythrum salicaria). The emergent
macrophytes were Lemna, wild celery and watercress. The submerged
macrophytes consisted of pondweed, moss and Callitriche.
The quality of the Mullaghanoe River has varied considerably over the last
two decades. It fell from a Q4 in 1983 to a Q3 in 1989 and rose again to a
Q4 in 1993. By 1995 the quality had declined again and it was assigned a
Q2-3 which it maintained until it rose again to a Q3 in 2000. This river
becomes very enriched and contained sewage fungus and solids quite a few
times during the study programme.
Fig. 1.10 outlines the catchment characteristics of the Mullaghanoe River
showing the CORINE 2000 landuse, GSI rock units (bedrock geology) and
the river network indicating Strahler stream order
41
#
#
#
##
0 3 6 Kilometers
N
EW
S
0 3 6 Kilometers
N
EW
S
Fig. 1.10 Catchment characteristics for Catchment of the Mullaghanoe River Sample Site 34M030100
showing CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler
stream order.
Sampling location
42
1.7.10 Site 10: The Mad River
The Mad River flows through the village of Cloonacool, which is
approximately 6km west of Tobercurry in Sligo. The study site (EPA
national river code 34M04-0100), was located approximately 300m
upstream of the bridge in Cloonacool. Monthly sampling was carried out in
an area measuring 4m x 15m. Its co-ordinates are E: 149230, N: 317304;
Latitude: 54° 06’ 07’’ N, Longitude: 8° 46’ 23’’ W. The Mad River is a
third order river with a catchment area of 7.77 km2. Pasture is the
predominant land use and there is a large forestry plantation located
upstream of the study site.
The average wet width in the study area was 4-6m during the sampling
period. Boulders and cobble dominated the substratum type with a
moderate degree of siltation. The dominant tree species present were
hawthorn, willow, ash, sycamore, alder and Gorse. During the course of this
study there were very few macrophytes found in the river. Moss and
liverworths were the most common submerged plants. The bankside
vegetation consisted of rush and Petasites spp.
Owing to the steep gradient of this mountain river it is prone to flash floods.
Sheep graze in upstream pasture where overgrazing is evident and cattle
have access to the river near the bridge for drinking purposes. The
macroinvertebrate diversity has diminished in recent years with a notable
absence of Ecdyonurus spp., Perla spp. and Gammarus spp. In 1989 it was
assigned a Q5 and from 1993 to the present study, the river has fluctuated
between a Q3 and a Q3-4.
Fig. 1.11 outlines the catchment characteristics of the Mad River showing
the CORINE 2000 landuse, GSI rock units (bedrock geology) and the river
network indicating Strahler stream order
43
0 3 6 Kilometers
N
EW
S
0 3 6 Kilometers
N
EW
S
0 3 6 Kilometers
N
EW
S
Fig. 1.11 Catchment characteristics for Catchment of the Mad River Sample Site 34M040100 showing
CORINE 2000 landuse GSI rock units (bedrock geology) and river network indicating Strahler stream
order.
Sampling location
44
Chapter 2
Experimental enrichment of a high status river in the West of
Ireland: Effects of nutrient manipulation on the genus
Ecdyonurus and benthic chlorophyll levels.
2.1 Introduction
This study was undertaken in response to a perceived eutrophication
problem in Irish rivers and to investigate the effect of artificial
eutrophication on an oligotrophic system. Taxa such as the key indicator
genus Ecdyonurus are important in the Irish Water Quality Rating system
and this study investigates its sensitivity to eutrophication. The approach
adopted was to artificially increase the concentration of nutrients in a river
and observe the response of Ecdyonurus using a novel split-stream system
devised specially for this project. The experiment was predicated on the
basis of a phosphorus limited river system.
The most significant threat to the quality of freshwater in Ireland is
eutrophication. This is defined as the enrichment of waters by the nutrients
phosphorus and nitrogen, beyond natural levels (Bowman and Clabby,
1998). The response of lake phytoplankton and periphyton to nutrient
addition during the eutrophication process is well known (Lund, 1969;
Kalff, et al., 1975; Schindler, 1975, 1985; Stockner and Evans, 1974).
However, the changes in algal production and species composition
following nutrient addition in streams and rivers are not fully understood or
adequately documented (Stockner and Shortreed, 1978). There has been a
substantial amount of work published on the limiting role of nutrients in
primary production and detrital decomposition of lake and marine systems.
However, efforts to understand their role in running waters has been a much
slower process (Elwood et al., 1981). Despite the fact that detrital
decomposition and algal production are known to be influenced by both
nitrogen and phosphorus (e.g., Hynes and Kaushik, 1969; Rhee, 1978;
Smith, 1979), there is little known about the mechanisms which drive
nutrient limitation in natural streams.
45
Wetzel (1983) made comparisons between the relative amounts of different
elements required for algal growth in freshwaters, emphasising the
importance of phosphorus and nitrogen. The ratio derived for phosphorus
was 80,000 while that of nitrogen was 30,000 (oxygen and hydrogen both
have a ratio of 1). This compares with, for example, a ratio for carbon and
silicon of 2,000 and 3,000, respectively. He pointed out that eventhough
variations in conditions of solubility or availability may at times make
abundant elements (e.g., silicon, iron and certain micronutrients), almost
unattainable, phosphorus and secondarily nitrogen, are generally the first to
impose limitation on the system.
Experiments on whole ecosystem enrichment have been well documented
and have been used to improve our knowledge and understanding of the
response of an ecosystem to man-made perturbation (Lock et al., 1990). In
lake systems, this approach was pioneered by Schindler and his fellow
workers (Schindler, 1985) at the experimental lakes in Canada. The levels
of phosphorus, nitrogen, sulphate, organic carbon and pH were manipulated
and findings showed that alterations of any one of these parameters had
detectable consequences on the other. The most notable change occurred
with the introduction of phosphorus. Primary production of phytoplankton
standing crop was stimulated in all cases, indicating that phosphorus was a
limiting nutrient in these oligotrophic lakes (Lock et al., 1990). Predicting
quantitative algal responses to increased loadings of nutrients in river and
stream ecosystems however is more difficult as they are in a constant state
of flux, both spatially and temporally (Bothwell, 1985). Extensive
laboratory research on P-limited growth kinetics of unicellular planktonic
algae has clearly shown that steady state growth rates saturate at extremely
low levels of ambient dissolved phosphorus (<10µgP/l) (Fuhs, 1969; Rhee,
1973; Tilman and Kilham, 1976; Brown and Button, 1979).
Natural phytoplankton populations contain carbon, nitrogen and phosphorus
in an C:N:P molar ratio of 105:15:1, and on average 1µg phosphorus
supports 1g chlorophyll production (Redfield, 1958; Ryther and Dunstan,
1971). Other studies on enrichment or eutrophication of running waters
46
have shown increases in periphytic biomass under certain conditions (e.g.
Elwood et al., 1981; Stockner and Shortreed, 1976; Horner and Welch,
1981; Horner et al., 1983; Perrin et al., 1987). Peterson et al. (1985)
reported the findings of a phosphorus enrichment experiment in a Tundra
river. He showed that an increase of only 10µg/l PO4-P above the
background (1-4µg/l), either alone or in combination with NO3-N, resulted
in substantial increases in epilithic chlorophyll a, metabolic activity (heat
output), bacterial activity and growth rates of two dominant riffle insects.
This study also revealed a decrease in species richness of diatom population
in response to enrichment.
Work carried out by Wuhrmann and Eichenberger (1975) seems to support
this conclusion. However, many workers have found that much higher
levels of phosphorus (>20µg/l PO4-P) are required to produce algal bloom
problems in streams and rivers (MacKenthun, 1968; Wong and Clark, 1976;
Horner et al., 1983). Elwood et al. (1981) found that although biofilm
biomass (grown on glass slides) in the phosphorus enriched section of a
second-order woodland stream was consistently higher than in the
unenriched section, only a small initial increase occurred in biofilm
chlorophyll a in situ at the enriched site. Observations showed that the
nitrogen-fixing cyanobacterium Nostoc sp. was significantly higher in the
enriched sections than the control section of the stream. This suggests the
operation of a dual N and P limitation effect in this stream (Lock et al.,
1990). A seasonal pattern in benthic algal biomass development is often
restricted to lowland streams with longer periods of stable discharge and
low-to moderate water velocity (Kjeldsen, 1996). In high gradient rivers
and streams with frequent changes in discharge, biomass development is
more dependent on the interval since the last flood event than on the season
(Tett et al., 1978; Fisher et al., 1982; Biggs and Close, 1989; Scarsbrook
and Townsend, 1993).
The impact of eutrophication on sensitive indicator taxa like Ecdyonurus
has not been investigated in detail to date in Ireland. This study examines
the effects of nutrient manipulation on Ecdyonurus venosus in two high
status rivers in the West of Ireland.
47
2.2 Study outline
Eutrophication is the main cause for the decline in water quality in Irish
rivers. There is a considerable lack of understanding in what controls the
disappearance of this sensitive indicator species as eutrophication impacts
on a river system. This investigation attempts to refine our knowledge of
the detailed mechanisms a river experiences as it departs from its pristine
state. It was hypothesised that the addition of orthophosphate to one section
of an oligotrophic river would cause an increase in the algal biomass
thereby affecting the distribution of the genus Ecdyonurus within the
experimental area.
48
2.3 Material and Methods
The first study was carried out in the Clydagh River during the summer of
2002 and repeated again during the summer of 2003. On completion of the
experiment in the Clydagh River at the beginning of July 2003, the split-
stream structure was moved to the Castlebar River. In total, the split-stream
experiment was carried out on three separate occasions.
2.3.1 Site 1: The Clydagh River
The Clydagh is a small river located 10km north of Castlebar town,
Co.Mayo. The study site (EPA national river code 34C05-0030) was
located at Latitude 53° 53’ 20” N: Longitude 9° 15’ 40” W. Discharge was
variable in this river: Flow patterns were “flashy” and corresponded rapidly
to precipitation. The mean flows in the river during the summer of 2002
and 2003 were 0.05 m3/s and 0.03 m
3/s, respectively. The site had an
average wet width of 7m during both sampling periods. The experimental
section was uniform with respect to gradient, substrate and canopy cover.
The bottom substrate consisted of a mixture of pebbles, cobble and large
boulders. The principal land use was predominately pasture with forestry
located upstream. The dominant riparian tree species were willow (Salix
spp.), hawthorn (Craetagus sp.) and birch (Betula pendula) while Fontinalis
spp., and other mosses were the submergent macrophytes. Sparganium
erectum was the predominant emergent plant.
2.3.2 Site 2: The Castlebar River
Details of the Castlebar River are outlined in Chapter 1, section 1.7.2.
Discharge in the river was regular and the mean flow during the study
period was 0.02 m3/s.
49
2.3.3 Preliminary investigations
Prior to setting up the split-stream experiment, an initial investigation into
the N:P ratio of eight potential river sites was undertaken. The Clydagh
river was sampled on 10/07/02 revealing a molar N:P ratio of 19; (N:P>15,
P is assumed to be limiting; N:P<15, N is assumed to be limiting). The
background orthophosphate concentrations were analysed in the form of
unfiltered Molybdate Reactive Phosphate (MRP µ/L P) and concentrations
were low (0.012-0.05 µ/l P). Other water quality parameters were also
examined throughout the study (see Table 2.1).
A three-minute kick sample for macroinvertebrates was taken on the same
day to establish the biological quality of the river. The Q-value of the river
was 4-5 with abundant numbers of the genus Ecdyonurus present. The
Clydagh River was chosen as the site for the split-stream experiment due to
suitable background water chemistry (P limited) and the presence of the
genus Ecdyonurus.
The experiment was replicated in the Castlebar River during the summer of
2003. The river was biologically and chemically assessed displaying a high
Q-value of 4-5, with abundant numbers of the genus Ecdyonurus present,
meeting the requirements for good water quality (low MRP concentration).
Both rivers were chosen as appropriate reference condition sites for the
nutrient manipulation experiments.
50
2.3.4 Experimental design and nutrient addition
2.3.4.1 Split-stream construction
The objective of the design was to split the river longitudinally into two
equal sections using a clear perspex barrier. This would prevent the river
water mixing on each side of the barrier. One side of the stream was to
receive an input of a known concentration of nutrient to continuously enrich
this section (treated section) while the other side was the untreated control
section. Installation of the split-stream barrier commenced on the 16th
July
2002 in the Clydagh River.
The study site was selected in a straight stretch of the river just downstream
of the bridge where a 10m long clear perspex barrier was constructed down
the centre of the river. Clear perspex sheets (2m x 0.5m x 5.5mm) were
chosen to avoid creating shadows on the riverbed or water surface thereby
introducing potential shading factors that could effect periphyton growth.
Angle irons (0.5m) were driven into the substratum until vertically stable at
1m intervals. It was essential that the perspex was flush with the riverbed
or, if possible, driven 4-5cm into the substratum. This was achieved firstly
by creating a longitudinal channel between the angle irons and then pushing
the perspex down into the river substrate.
Five perspex sheets were fixed in series to the iron bars using stainless steel
bolts. Once in place the 10m structure was supported vertically allowing the
natural free movement of water on either side of the barrier. The barrier
was fitted to the riverbed and any gaps were sealed with sand and small
pebbles to prevent mixing of water between the left and right hand sides of
the river. The river was left to settle for 3 weeks prior to the
commencement of the sampling programme. This enabled the river to
readjust to the disturbances it experienced during the instillation of the
barrier. To test the barrier for any leaks and to verify rapid homogenous
mixing fluorescent dyes (Rhodamine-B and Fluorescein disodium salt) were
then added to the experimental section. Fig. 2.1 and Fig. 2.2 show the
detailed layout of the split-stream sites.
51
Fig 2.1 Detailed view of the split-stream structure in the Clydagh River
during 2002 and 2003.
Fig 2.2 Detailed view of the split-stream structure in the Castlebar River
during 2003.
3m width
10m Perspex
barrier
Dye trace to test for
leaks in barrier
“treated”
section
untreated
“control”
section Nutrient
input site
25m
Perspex
barrier
Fluorescein
disodium salt
dye trace
untreated
“control” section
“treated”
section
Nutrient
input site
1-2m width
52
2.3.4.2 Nutrient addition
Based on previous studies carried out by Peterson and his co-workers in
1985 who studied the effects of nutrient enrichment on a Tundra River in
northern Alaska over a 6 week period, it was proposed to increase the
background orthophosphate concentration of the river water by 10µg/l in an
attempt to artificially enrich one side of the split-stream. Such a low level
also ensured that the river was not ‘polluted’ or over enriched by the
experiment. On each site visit the flow in the manipulated section was
measured in m3/s. The nutrient solution was made up to a concentration of
20g P/l using potassium dihydrogen orthophosphate salt (KH2PO4) which
was stored in a 20litre rectangular polycarbonate carboy. This was placed in
a steel box at an elevated position on the bank of the river. The water
chemistry of the river in its natural state was analysed 1-2 days before each
visit thereby supplying information on the background concentration of
orthophosphate in the river. Depending on the flow in the river at each site
visit, the nutrient stock concentration was adjusted accordingly, thereby
delivering a continuous supply of P, in order to raise the river water
concentration by 10µg/l. The target concentration was achieved using the
following mass balance equation:
Concentration downstream = (f x c + F x C)/(f + F)
f = flow in peristaltic pump (l/s)
c = concentration of the nutrient stock solution in mg/l
F = Flow in the experimental channel being dosed (l/s)
C = assumed upstream concentration of P in mg/l
The solution was added continuously to one side of the river using a gravity
fed Mariotte siphon feeding system at a rate of approximately 1ml per
minute. The site was visited 2-3 times per week to check flows, carry out
routine sampling and change the battery supplying the pump. This close
monitoring of the system ensured a continuous supply of nutrient to the
manipulated section at the correct concentration. The physico-chemical
parameters listed in Table 2.1 were monitored at various intervals within
each side of the split-stream on a weekly basis. Water samples were taken
at the upstream end of each side of the split-stream (0m at nutrient input and
53
0m control section). Depending on the experiment, the water was analysed
at 10m and 20m distances from the top end in both sides and again upstream
and downstream (20m, 30m) of the experimental site.
The Mariotte siphon feeding system was used to deliver nutrient to the
Clydagh River in 2002 over a period of 6 weeks. In June 2003, the
experiment was repeated by manipulating the other side of the river. The
nutrient was released into the opposite side for the duration of 5 weeks using
a battery powered peristaltic pump, which provided a more reliable nutrient
delivery system. The same system was used again when the experiment was
moved to the Castlebar River in July 2003 and nutrient was added over a 9
week period. The split-stream structure was extended in this river by a
further 10m to 20m to allow for increased sampling within the study area
and an improved opportunity to assess the response over distance from the
nutrient delivery point.
2.3.4.3 Chlorophyll a estimations
Algal biomass, was estimated using chlorophyll per unit area measurements
obtained from scrapings taken from the stone surfaces in the river. The
standard hot methanol chlorophyll a extraction method adopted by the EPA
was used to estimate photosynthetically active biomass on the surfaces
sampled. Within each of the “control” and “treated” sections, riffled areas
were selected for sampling. All stones sampled were covered by at least 20
cm of water for 4 weeks before sampling commenced. Using a stiff
toothbrush, a 10cm2
section of stone surface was scraped from 5 stones in
series within the first 2m of each side of the experimental site. These were
then pooled into a 500ml container of river water. At 2m intervals along the
10m longitudinal gradient, this process was carried out until a total of 5
pooled samples were gathered from both the treated (experimental) and
untreated (control) sides of the split-stream (see Fig. 2.3). Samples were
taken back to the laboratory and analysed on the same day. Chlorophyll a
concentrations were measured once prior to the addition of nutrient and
again over a period of weeks (depending on experiment) during the nutrient
manipulation experiment.
54
As previously mentioned, the split-stream structure was extended in the
Castlebar River in 2003 from 10m to 20m. During the last three weeks of
the programme in this experiment the sampling regime was intensified.
Using random numbers in a grid-like representation of the experimental site,
25 stone scrapings were taken separately along each side of the 20m split.
The flow of the river water was concentrated along the spilt-stream
structure, particularly in low flows so it was decided to take algal scrapings
randomly, 15cm and 30cm from the perspex barrier (see Fig. 2.4). The
chlorophyll a concentrations were measured individually instead of
measuring 5 pooled samples, in order to improve the power of the statistical
analyses. In the Clydagh River during the summer of 2002, algal biomass
estimations were taken on four separate occasions over a 6-week period. In
the summer of 2003, algal biomass scrapings were measured on nine visits
over a 5-week period. Finally, in the Castlebar River in 2003, ten algal
measurements were established over a period of 8-weeks.
55
x x x 0m
x
x
x x 2m
x x
x
x x 4m
x x
x
x x 6m
x x
x
x x 8m
x x x
x x 10m
x x
x
x x x 0m
x
x
x x 2m
x x
x
x x 4m
x x
x
x x 6m
x x
x
x x 8m
x x
x
x x 10m
x x
x
Fig 2.3 Sampling regime in the Clydagh River in 2002 and 2003.
25m perspex barrier
0m x
x
x
x
5m x x x
x
x
10m x x
x
x
x
15m x x
x
x x
20m x x
x
x x
25m x
x 0m
x
x
x
x 5m x x x
x
x 10m x
x x
x
x x15m x
x
x
x x 20m x
x x
x 25m
Fig 2.4 Sampling regime in the Castlebar River in 2003.
5 scrapings
pooled into 1
sample every
2m along the
10m gradient
Direction of
water flow
10m perspex
5 scrapings
pooled into 1
sample every
5m along the
25m gradient
Direction of
water flow
Direction of
water flow
Samples taken
15cm from the
barrier
Samples taken
30cm from the
barrier
56
2.3.4.4 Macroinvertebrate sampling
Quantitative sampling, using a 36cm x 36cm metal-framed Surber sampler,
was carried out prior to the addition of the nutrient and on subsequent visits
during the nutrient manipulation experiment. In the Clydagh River during
the summer of 2002, macroinvertebrates were sampled on four separate
occasions over a 6-week period. No macroinvertebrates were sampled in
the Clydagh River in 2003 due to low numbers of the Ecdyonurus in the
river. In the Castlebar River in 2003, macroinvertebrate samples were taken
on seven occasions over a period of 9 weeks.
Sampling was initiated within riffled sections of the split-stream sites and 5
Surbers were taken on either side of the split-stream during each sampling
trip. The Surber was placed on the riverbed facing upstream and the
substrate within the metal frame disturbed. All macroinvertebrates were
dislodged into the open net and special care was taken when collecting the
Ecdyonurus specimens. The samples were preserved in 70% IMS on site
for later identification. As with the collection of stone scrapings, Surber
sampling was carried over a period of weeks during the nutrient
manipulation experiment (time scale depending on particular experiment).
The macroinvertebrate Surber samples were taken at 2m and 5m intervals
(depending on the experiment) along the longitudinal gradient of the split-
stream barrier (Fig 2.3 and Fig 2.4).
Comparisons between periphyton biomass and macroinvertebrates could
then be made between the treated and control sections. All
macroinvertebrates were counted and identified to family level while the
Ecdyonurus specimens were identified to species level.
57
2.4 Results
Results are shown comparing chlorophyll levels in the treated and control
sides of the split-stream. Macroinvertebrates are also compared on either
side. The minimum, mean, median, maximum and standard deviation
values for the background physico-chemical parameters measured during
the three experiments are outlined in Table 2.1. As indicated above,
phosphate was added in order to raise the ambient concentration in the
treated side of the stream by 10µgP/l when mass balanced. Water samples
also collected for physico-chemical analysis immediately downstream of the
split-stream during the enrichment process and at 20m, 30m, 50m, 100m
downstream showed no detectable downstream gradients in the MRP
concentrations.
The effects of nutrient addition on chlorophyll a (µg/cm2) concentrations
and macroinvertebrates between the treated and control sections during the
three experiments were examined using a repeated measures analysis of
variance (ANOVA). Statistical analysis was performed using SPSS version
11.0 software. This analysis examines the temporal variation of the
chlorophyll a and macroinvertebrate Surber samples along the gradient of
the split-stream barrier (positional information) during the course of the
experiment.
The general linear model was used to analyse the data. In SPSS, the
Within-Subjects Variables correspond to the number of weeks the
experiment was carried out for. The Between-Subjects Factors relates to the
“treated section” (manipulated) and the “control section” (non-manipulated)
of the experimental divide. The positional information (distance from the
top to the bottom of the perspex barrier) representing the spatial variation
along the gradient during the course of the experiment are specified using
the covariance structure for the residuals.
Results are outlined in Table 2.2. Chlorophyll a (µg/cm2) and
macroinvertebrate data were transformed as needed to stabilise variance and
to improve normality.
58
Table 2.1 Maximum, minimum, mean, median and standard deviation values for the physico-chemical parameters examined in the Clydagh River in
2002, 2003 and in the Castlebar River in 2003 during the split-stream experiments.
Temperature
°C
DO %
Saturation
pH
pH units
Conductivity
µS/cm
Orthophosphate
mg/l P
TON
mg/l N
Ammonia
mg/l N
Chloride
mg/l Cl
Alkalinity
mg/l
CaCO3
BOD5
mg/l O2
Colour
Hazen
units
Clydagh River 2002
n 6 6 12 12 12 12 12 11 12 10 11
Minimum 10.3 97.0 7.4 68.0 0.012 0.05 0.005 11 14.0 0.40 33.0
Mean 13.3 102.0 7.9 172.0 0.027 0.05 0.007 13.7 60.5 0.79 141.8
Median 13.9 100.5 8.1 164.0 0.025 0.05 0.005 14 54.0 0.80 110.0
Maximum 15.7 109.0 8.2 276.0 0.050 0.05 0.02 16 116.0 1.00 309.0
Standard
deviation
1.9 4.9 0.2 68.9 0.012 1.12E-09 0.005 1.7 34.9 0.22 81.0
Clydagh River 2003 n 14 14 14 14 15 15 15 15 11 7 13
Minimum 5.8 94.0 7.1 87.0 0.012 0.05 0.005 14.0 10.0 1.00 26.0
Mean 12.9 102.7 8.1 228.8 0.021 0.05 0.005 17.7 87.1 1.01 71.6
Median 13.4 102.0 8.2 255.0 0.019 0.05 0.005 17.0 92.0 1.00 54.0
Maximum 18.4 111.0 8.3 287.0 0.030 0.05 0.005 21.0 120.0 1.10 153.0
Standard
deviation
3.7 4.7 0.3 59.7 0.006 7.04E-10 8.8E-11 2.21 32.9 0.04 40.9
Castlebar River 2003 n 15 15 16 18 18 18 18 18 9 5 10
Minimum 9.6 96.0 7.1 83.0 0.028 0.05 0.005 13.0 14.0 1 40.0
Mean 13.9 102.9 7.9 197.6 0.036 0.06 0.005 20.6 52.0 1 110.2
Median 14.0 103.0 7.9 220.5 0.037 0.05 0.005 21.0 52.0 1 94.5
Maximum 18.9 112.0 8.3 252.0 0.044 0.155 0.005 25.0 82.0 1 220.0
Standard
deviation
2.9 4.17 0.3 49.9 0.004 0.02 5.64E-11 2.5 26.1 0 59.5
59
Table 2.2 Repeated measures analysis of variance of chlorophyll a, Ecdyonurus genus numbers and total macroinvertebrate numbers.
River/Parameter
Source of variation
(between groups) Source of variation
(within-subjects effects)
Clydagh River 2002
Treated vs control section Gradient
Week Week * Gradient
Week * treated vs
control section
Chlorophyll a
F1,7 = 0.004
P = 0.950
F1,7 = 2.48
P = 0.159
F 3,21 = 2.153
P = 0.124
F 3,21 = 0.901
P = 0.457
F 3,21 = 0.345
P = 0.793
Ecdyonurus genus numbers/m2
F1,7 = 0.029
P = 0.869
F1,7 = 1.507
P = 0.259
F3,21 = 2.190
P = 0.119
F3,21 = 1.370
P = 0.279
F3,21 = 0.654
P = 0.589
Total number macroinvertebrates/m2
F1,7 = 2.113
P = 0.189
F1,7 = 2.877
P = 0.134
F3,21 = 3.569
P = 0.031*
F3,21 = 1.316
P = 0.295
F3,21 = 0.158
P = 0.923
Clydagh River 2003
Chlorophyll a
F1,7 = 26.867
P = 0.001***
F1,7 = 6.725
P = 0.036*
F8,56 = 1.837
P = 0.089
F8,56 = 2.595
P = 0.017*
F8,56 = 4.290
P = 0.001***
Castlebar River 2003
Chlorophyll a
F1,7 = 18.630
P = 0.003**
F1,7 = 0.170
P = 0.692
F9,63 = 6.068
P = 0.001***
F9,63 = 0.538
P = 0.841
F9,63 = 2.674
P = 0.011*
Ecdyonurus genus numbers/m2
F1,7 = 5.414
P = 0.053
F1,7 = 0.384
P = 0.555
F6,42 = 0.499
P = 0.806
F6,42 =1.042
P = 0.412
F6,42 =2.005
P = 0.086
Total number macroinvertebrates/m2
F1,7 = 0.054
P = 0.823
F1,7 = 17.339
P = 0.004**
F6,42 = 4.092
P = 0.003**
F6,42 = 3.923
P = 0.003**
F6,42 = 2.990
P = 0.016*
p values:
* p<0.05; ** p<0.01; *** p<0.001; ns – not significant
The physico-chemical measurements outlined in Table 2.1 are average values of the “natural” background concentrations of the parameters in the river (upstream of split-
stream site) taken over the experimental periods. The N:P ratios of the rivers in their “natural” state are outlined in Table 2.3. N:P ratios were determined as TON (Total
Oxidised Nitrogen) + ammonia divided by the orthophosphate concentration in molar quantities.
60
Table 2.3 Variation in the background N:P molar ratios of the Clydagh River and the Castlebar River during the three experimental sampling
periods.
Date Clydagh
River 2002
Date Clydagh
River 2003
Date Castlebar
River 2003
10/07/02 19 13/03/03 9 08/04/03 4
24/07/02 6 20/03/03 8 23/04/03 4
20/08/02 3 26/03/03 6 08/05/03 4
22/08/02 3 03/04/03 9 19/05/03 4
29/08/02 5 08/04/03 8 10/06/03 3
04/09/02 6 23/04/03 9 30/06/03 3
12/09/02 8 19/05/03 10 07/07/03 3
19/09/02 5 05/06/03 6 14/07/03 3
02/10/02 5 16/06/03 4 29/07/03 4
03/10/02 6 19/06/03 4 24/07/03 4
09/10/02 6 23/06/03 4 28/07/03 4
15/10/02 5 26/06/03 5 31/07/03 4
30/06/03 5 05/08/03 3
03/07/03 5 07/08/03 3
07/07/03 6 11/08/03 3
14/08/03 3
20/08/03 3
09/09/03 8
17/09/03 9
61
2.4.1 Effects of nutrient enrichment on periphyton growth
2.4.1.1 Experiment 1 - Clydagh River: 26th
July - 15th
October 2002
On initial examination of the river in July 2002, an N:P ratio of 19 indicated
that the river was P-limited (Table 2.3). The N:P ratio however dropped
substantially over the remainder of the experiment to low levels that
suggested N limitation rather P limitation. There was no significant
difference in the chlorophyll a biomass between the “treated” section and
the “control” section of the spilt-stream experiment throughout the
experiment (Table 2.2). Algal standing crop did not respond to phosphorus
enrichment and biomass levels remained low during the entire experiment
(Fig. 2.5).
Fig. 2.5 Mean chlorophyll a concentration (µg/cm2) in the “treated” section
(+10µgP/l) vs the “control” section. Bars are means ± standard deviation: n
= 5.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
20/0
8/02
27/0
8/02
03/0
9/02
10/0
9/02
17/0
9/02
24/0
9/02
01/1
0/02
Ch
loro
ph
yll
a µ
g/c
m2
"Control" section
"Treated" section
62
2.4.1.2 Experiment 2 – Clydagh River: 5th
June - 11th
July 2003
The N:P ratios from March to July 2003 in the Clydagh River also remained
low throughout the entire study (Table 2.3). There was an overall
significant difference (Table 2.2) between the “treated” and the “control”
sides of the Clydagh River along the temporal gradient during the sampling
period. Visible sloughing of the periphyton mats began to occur after day
28. The chlorophyll a concentrations began to increase after day 32,
particularly in the P-treated section (Fig. 2.6).
Fig. 2.6 Mean chlorophyll a concentration (µg/cm2) in the “treated” section
vs the “control” section. Bars are ± standard deviation: n = 5. Negative
values have been omitted for clarity.
0
5
10
15
20
25
30
05/0
6/03
12/0
6/03
19/0
6/03
26/0
6/03
03/0
7/03
10/0
7/03
Date
Ch
lorop
hyll
a µ
g/c
m2 "Control" section
"Treated" section
63
2.4.1.3 Experiment 3 – Castlebar River: 18th
July – 12th
September 2003
The N:P ratios in the Castlebar River stayed low throughout the entire study
(Table 2.3). The mean chlorophyll a concentrations increased substantially
in the last three weeks of the study (Fig. 2.7) with a significant difference in
the periphyton biomass (Table 2.2; p=0.003) between the “treated” and the
“control” sections during the course of the study. There was also a
significant trend through time in the chlorophyll a concentration (Table 2.2:
p=0.001) with a quadratic fit to the trend line of increasing chlorophyll.
There were no significant differences in the chlorophyll a concentrations
along the gradient however indicating that even though there were
significant increases in the concentrations over time, chlorophyll a
concentrations did not increase along the length of the perspex barrier. In
other words, the concentrations were not higher at the bottom of the
experimental divide compared to the top i.e. from 0-20m).
Fig. 2.7 Mean chlorophyll a concentration (µg/cm2) in the “treated” section
(manipulated) vs the “control” section. Bars are means ± standard deviation:
n = 5. Negative values having been omitted for clarity.
0
50
100
150
200
250
21/0
7/03
28/0
7/03
04/0
8/03
11/0
8/03
18/0
8/03
25/0
8/03
01/0
9/03
08/0
9/03
15/0
9/03
Date
Ch
lorop
hyll
a u
g/c
m2
control section treated section
64
2.4.2 Effects of nutrient enrichment on macroinvertebrates
2.4.2.1 Experiment 1 - Clydagh River: 26th
July - 15th
October 2002
The response of nymphs of the Ecdyonurus genus to nutrient addition
showed no significant difference (Table 2.2) between the “treated” and the
“control” sections of the Clydagh River during the 6-week sampling period.
The graph in Fig. 2.8 shows the average numbers of Ecdyonurus genus per
m2 in both sections of the river over the experimental period. There was a
significant difference in the total number of macroinvertebrate fauna on the
temporal scale only (Table 2.2; p=0.031) and no other significant
differences were observed at any level. The total numbers of
macroinvertebrates per m2
are shown in Fig. 2.9.
Fig. 2.8 Ecdyonurus genus numbers per m2 in the Clydagh River during the
split-stream experiment in 2002. Bars are means ± standard deviation: n =
5.
0
10
20
30
40
50
60
70
26/0
7/02
09/0
8/02
23/0
8/02
06/0
9/02
20/0
9/02
04/1
0/02
18/1
0/02
Date
Ecd
yon
uru
s gen
us
nu
mb
ers/
m2
"Control" section
"Treated" section
65
Fig. 2.9 Total macroinvertebrate numbers per m2
in the Clydagh River
during the experiment in 2002. Bars are means ± standard deviation: n = 5.
0
100
200
300
400
500
600
700
800
900
1000
26/0
7/02
09/0
8/02
23/0
8/02
06/0
9/02
20/0
9/02
04/1
0/02
18/1
0/02
Date
Tota
l n
um
ber o
f m
acroin
verte
brate
s p
er
m2
"Control" section
"Treated" section
66
2.4.2.2 Experiment 2 - Clydagh River: 5th
June - 11th
July 2003
Weekly biological assessments were performed from mid-March 2003 to
July 2003 prior to commencement of the experiment. Results revealed a
marked absence of the genus Ecdyonurus during this time. Immature
juveniles of this genus were present in low numbers on occasions with
sporadic findings of mature larvae. This was in complete contrast to the
previous summer where Ecdyonurus genus was abundant. Due to the
“flashy” nature of the river, it is thought that eggs or the immature larvae
may have been displaced from this site during a flood in early
spring/summer (Appendix 5.3). For this reason, the effects of nutrient
manipulation on the macroinvertebrate fauna were not examined during this
experiment.
2.4.2.3 Experiment 3 – Castlebar River: 18
th July – 12
th September 2003
Here the macroinvertebrates were collected over a period of 7-weeks during
the nutrient addition experiment. The graph in Fig. 2.10 represents the
numbers of Ecdyonurus genus per m2 in both sections of the river over the
experimental period. The total numbers of macroinvertebrates per m2
in
both sides of the split-stream experiment are shown in Fig. 2.11. There was
no significant difference in numbers of Ecdyonurus genus (Table 2.2).
A significant difference in the abundance of macroinvertebrate fauna (Table
2.2; p=0.004) was observed along the gradient only and there was no
significant differences between the “treated” and the “control” sections per
se. Significant differences were found at all levels for the “within-subjects
effects” (Table 2.2).
67
Fig. 2.10 Ecdyonurus genus numbers per m2 in the Castlebar River during
the split-stream experiment in 2003. Bars are means ± standard deviation: n
= 5.
Fig. 2.11 Total numbers of macroinvertebrates per m2 in the Castlebar River
during the split-stream experiment in 2003. Bars are means ± standard
deviation: n = 5.
0
50
100
150
200
250
18/0
7/03
25/0
7/03
01/0
8/03
08/0
8/03
15/0
8/03
22/0
8/03
29/0
8/03
05/0
9/03
12/0
9/03
Date
Ecd
yon
uru
s n
um
bers
per m
2
"Control" section
"Treated" section
0
500
1000
1500
2000
2500
3000
3500
4000
4500
18/0
7/03
25/0
7/03
01/0
8/03
08/0
8/03
15/0
8/03
22/0
8/03
29/0
8/03
05/0
9/03
12/0
9/03
Date
Tota
l m
acroin
verte
brate
nu
mb
ers
per m
2
"Control" section
"Treated" section
68
2.5 Discussion
2.5.1 Introduction
While some of the nutrient manipulation experiments showed significant
differences between the control and treated sections, not all did so. In
addition, it is interesting to note that sites appeared to be N-limited but
seemed to respond to P, particularly after 6-7 weeks of nutrient enrichment.
The phosphorus enrichment experiments carried out in the Clydagh River in
2002 and 2003 and the Castlebar River in 2003 produced some interesting
and valuable insights into the ecology of reference condition river stretches.
Each of the three enrichment experiments are discussed individually and
then comparisons are made later with other studies from the literature.
2.5.2 General findings in the Clydagh River 2002
There was no significant difference in the growth of periphyton biomass in
the enriched section, compared with the untreated section of the Clydagh
River in 2002. In addition, there were no significant differences in the
numbers of Ecdyonurus genus between both sections during the study
period. Significant differences in total macroinvertebrate numbers were
observed over time only. Larval numbers increased in both sections of the
experimental divide from mid-September on. This period (September to
October) signals the hatching of eggs from the benthos into immature larvae
which is quite typical in the life cycle of many macroinvertebrates. No
significant differences were observed between both sections of the
experimental divide and the increase in the abundance of macroinvertebrates
more than likely corresponds to the natural changes that occurred in the life
cycles of the larvae which are unrelated to the experiment.
The N:P ratio of the Clydagh River prior to setting up the split-stream
experiment was 19. This was within the range of reported molar N:P ratios
for algae 15:1 (Redfield, 1958) to 30:1 (Rhee, 1978) and would include the
range of N:P ratios where dual N and P limitation would be expected.
Unfiltered MRP concentrations averaged at 27-µgP/l over the experiment
while the mean DIN concentration (NO3 + NO2 + NH3-N) was 57µgN/l. At
69
this average P concentration, small changes in the concentration of either
nutrient, particularly P, should have had a significant effect on the N:P ratio.
The background N:P ratios dropped to low levels two weeks later and
fluctuated between ratios of 3 and 8 as the experiment progressed until the
middle of October.
The mean chlorophyll a concentrations highlighted the lack of response to
phosphate addition. Concentrations remained similarly low on both sides of
the split-stream during this period. The algae in the river appear to be under
a dual N and P limitation as the N:P ratios fluctuate below or above the
growth optimum. Manuel and Minshall (1978) found significantly higher
levels of chlorophyll a and aufwuchs biomass in experimental streamside
channels enriched with N and P, using stream water with an ambient N:P
ratio of 12, a value at which N limitation would be expected. However,
there were no significant increases in chlorophyll a or aufwuchs biomass
when the stream itself was enriched in situ. These authors imply that
increased grazing effects and sloughing of algal biomass in the natural river
may have masked the effects of enrichment on the periphyton growth.
The low N:P ratios coupled with increased grazing by macroinvertebrates
may have contributed to the lack of apparent algal response to enrichment in
the present study. Unfortunately it was not possible to measure grazing
pressures, an area which should be examined in future investigations. It is
also possible that algal seed populations did not establish themselves during
early spring 2002 or may have been depleted due to floods during this
period (Appendix 5.3) caused by the “flashy” nature of this river. There
may also be fluxing of nutrients from the forestry plantation, septic tanks
and farms upstream of the site causing the fluctuations in the N:P ratio. The
N:P ratios were only measured on approximately a weekly basis over a three
month period so it would be important to monitor these ratios on a weekly
basis over the course of an entire year.
70
2.5.3 General findings in Clydagh River 2003
The experiment was repeated in the Clydagh River in 2003, but the nutrient
was applied to the opposite side of the river for this trial. There was a
significant difference in the growth of periphyton biomass (chlorophyll a
µg/cm2) in the enriched section, compared with the untreated section. The
average N:P ratio from March to July 2003 was 6, reaching a high of 10 in
May 2003, appearing to indicate an N limited system. The mean
chlorophyll a concentration in both sides of the river showed an increase in
the periphyton biomass particularly during the last two weeks of the
experiment. It is difficult to say what may be driving the growth of the
periphyton during this experiment, when there was no significant difference
in biomass levels during the experiment in the same river in 2002 and the
fact that the river displayed such a low N:P ratio. There may be a year
effect driving the change or differences found between both the experiments
carried out in the Clydagh River (2002 and 2003). During the study visual
observations suggested that algal growth appeared to be at its greatest
towards the lower end of the split-stream (10m long). This suggested that
uptake of the nutrient was not immediate and occurred further downstream
of the point source. There is the possibility that there was fluxing of N
through the river system that was not detected during the sampling
programme. There may be nutrients pulsing from the forestry plantation
upstream during rainfall events, causing intermittent releases of N into the
system, thereby fluctuating the N:P ratios between sampling periods. This
may have contributed to an increase in periphyton growth in the enriched
section of the split-stream towards the end of the experiment but must
remain as speculation. High resolution time-proportional sampling designed
to pick up diurnal or day to day variations in N:P would be required to
verify this hypothesis.
A biological assessment was carried out on a weekly basis (March – June
2003) prior to manipulation of the nutrient. The river did display a high Q-
value of 4-5 but there was a marked absence of the genus Ecdyonurus
during this period with the exception of the small numbers of immature
juveniles that were periodically found. The abundance of Ecdyonurus in the
same river was far greater during the summer of 2002 in comparison to the
71
summer of 2003. As previously mentioned, this river is prone to floods and
due to the scouring nature of the substrate, it is thought that the eggs or the
immature larvae of the genus Ecdyonurus may have been dislodged during a
flood in the spring 2003. Because of this, the effects of nutrient enrichment
on the macroinvertebrate fauna were not examined during this experiment.
2.5.4 General findings in the Castlebar River 2003
The experiment was replicated by moving it to the Castlebar River and the
length of the split-stream was extended to 25m. A significant difference in
chlorophyll a concentrations was detected over time between the control
section and the treated sections of the experiment. The N:P ratios monitored
from April to September 2003, gave a mean N:P ratio over this period of 4.
As with the Clydagh River, this indicated an N-limited system.
Interestingly, the highest significant difference in the chlorophyll a biomass
levels between the enriched and control sections throughout all three
experiments were found in this river. The lowest N:P ratios were also
measured in this system. There was no significant difference in Ecdyonurus
genus numbers per m2 between either side of the split-stream. A significant
difference in the total number of macroinvertebrates was found over time
along the gradient but was not observed between the control and the treated
sections of the experiment per se. As with the Clydagh River in 2002, the
significant difference in macroinvertebrate abundance during the course of
the experiment probably reflects the natural changes that occur in rivers
during emergence and hatching processes.
After 6-weeks, the sampling regime for estimating periphyton biomass was
modified to measuring individual stone scrapings as opposed to measuring a
pooled sample of 5 scrapings over the 20m. This improved the power of the
sampling regime and the statistical analyses. The chlorophyll a levels began
to increase after 6-weeks of continuous nutrient enrichment. Algal standing
crop appeared to respond to the phosphorus enrichment and biomass levels
increased steadily during the last three weeks of the experiment. This was
also reflected in the chlorophyll a concentrations which also showed a
72
significant trend through time with a quadratic fit to the trend line of
increasing chlorophyll.
As in the Clydagh River, there is the possibility of the fluxing or pulsing of
N through the river system, which are not detected between sampling
periods. Domestic houses in close proximity to the river could possibly be
introducing N sporadically into the river. Potential upstream sources
include a concrete manufacturer, flushing of toilets by night, overflowing
septic tanks or from waste grey water coming from washing machines and
drains nearby. It appears that enough N is being released into the river e.g.
from septic tank hydraulic pressure in the evenings, to allow all the
background P to be taken up, plus the additional 10µgP/l released into the
river via the nutrient manipulation experiment. The addition of N to the
system in this manner may be contributing to the increased algal growth in
the enriched section, but again, must remain as speculation.
73
2.5.5 Comparisons with other studies
Determining the likely biological impact of nutrient enrichment is especially
complicated for flowing systems (van Nieuwenhuyse and Jones, 1996). The
recognised need to include the influence of processes that occur upstream
led to the development of the river continuum theory (Vannote et al., 1980).
An improved understanding of the temporal aspects of nutrient loss with
those of biological demand is necessary, particularly in view of the
possibility of a seasonal dependence of N limitation (Axler et al., 1994;
Peterson et al., 1997). Studies by Heathwaite et al., (1996) demonstrated in
terms of nutrient loss, a good correlation between total area of agricultural
land and river nitrate concentrations. Meyer et al., (1988) summarised into
three areas, the reasons why a greater understanding of elemental cycling
within lotic systems is necessary to improve understanding of ecosystem
behaviour. These were: (1) nutrients regulate ecological processes in
streams, especially when a nutrient is limiting; (2) nutrients link terrestrial
and aquatic ecosystems: and (3) stream processes alter the timing,
magnitude and form of elemental fluxes and therefore nutrient availability
to downstream communities (Edwards et al, 2000).
Studies have also been carried out in which stimulatory effects of
enrichment have been observed (Elwood et al., 1981). Work carried out by
Wuhrmann and Eichenberger (1975) revealed no significant effect of P, N
or N+P enrichment on periphyton growth receiving groundwater with a
concentration of 10µgP/l. The N:P ratio in the stream channels was 96:1, a
value where P-limitation would be expected. In a second experiment, a
significant increase in biomass was observed when a mixture of trace
elements was added to the stream water (Elwood et al., 1981). It may be
possible that the primary producers were limited during this period by one
or more essential elements but these were not examined in the experiments.
Statistical analysis indicated there were no significant differences in
macroinvertebrate numbers per m2 and Ecdyonurus numbers per m
2
between both sides of the split-stream in the Clydagh River, suggesting
there was equal grazing pressure on either side of the experimental divide.
74
2.5.6 N:P Ratios
The vast majority of rivers in Ireland are P limited and an analysis of the
EPA database of the N:P ratios for 99 rivers in the West of Ireland found
that approximately 4% of samples analysed are N-limited and with low
MRP concentrations (<0.05 mg/l P; Appendix 5.14) so it was assumed that
the river systems chosen for these experiments would also be P limited. The
Clydagh River was predicated a P-limited system on the basis of displaying
an N:P ratio of 19, measured in July 2002. The ratio dropped considerably
thereafter and remained low when the experiment was repeated during the
following summer. The study was replicated in the Castlebar River during
the summer of 2003 but analysis of the water revealed yet again another
apparently N limited site. In view of the fact that only a small percentage
of rivers in Ireland are N limited, the dynamics of these high status rivers
are interesting from an ecological point of view. It was envisaged that by
artificially enriching a P limited river, algal growth would increase and the
effects on the sensitive indicator genus Ecdyonurus could be described.
This genus was very abundant in the Castlebar River, even though the
phosphorus concentrations exceeded the annual median 30µgP/l
concentration set in the Phosphorus Regulations (DELG, 1998) on many
occasions during the study although it was always satisfactory from a
biological quality point of view. The mean background MRP concentration
in the Castlebar River during the study in 2003 was 37µgP/l. At an N:P
ratio of 4, it seems likely that the system was N-limited at the point when
the samples were taken. Nonetheless, the fact that P additions apparently
stimulated growth suggests that if nitrogen becomes available in the system,
the available P is rapidly depleted and that the added 10µgP/l was also
consumed. However, the river has a Q-value of 4-5 with low levels of algal
growth.
In a study carried out by Edwards et al. (2000), the variations in nitrogen
dynamics of an upland river in northeast Scotland were examined. He
concluded that the source of nitrogen appeared to vary seasonally where it
originated primarily from in-stream biological production during the
75
summer months while in the winter-spring period leaching from the plant-
soil system appeared to be the major contributor. He found that on any
individual sampling day, a wide range of N:P ratios occurred in the
catchment studied. He observed the lowest N:P ratios during the summer
and early autumn, particularly for upland catchments dominated by semi-
natural vegetation. In many upland situations, physical scouring of stream
systems during spates, or chemical-induced inhibitions by high molecular
weight organic matter (Freeman et al., 1990), would be expected to limit
algal productivity at certain times of the year (Twist et al., 1998). Findings
from the Irish river nutrient enrichment experiment, clearly showed that
both rivers during the three experiments were N limited throughout the
summer/autumn months, as also found in studies carried out by Edwards
and his co-workers (Edwards et al., 2000).
76
Chapter 3
Investigations into the feeding regime of the mayfly
Ecdyonurus venosus.
3.1 Introduction
This study describes the feeding regime of the mayfly larvae Ecdyonurus
venosus by examining the gut contents of specimens taken from the
Castlebar River in the West of Ireland. This same river was used for the
split-stream experiment (2003) outlined in Chapter 2. The gut contents of
Ecdyonurus venosus larvae obtained from this river during the split-stream
experiment were investigated to establish the diet of these mayflies. The
genus Ecdyonurus is particularly important as an indicator species in the
ecological assessment of river water quality in Ireland due to its sensitivity
to pollution and widespread occurrence in Irish rivers (McGarrigle et al.,
1998). To date, there have been no studies carried out on the diet of this
species. It was hypothesised that the gut contents of the larvae may reflect
the changes in the algal biomass exhibited in the experimental site and
possibly reflect a variation in algal taxa between both sections of the
experiment – i.e. its feeding regime would change during the nutrient
manipulation experiment. The objective was to establish what these key
indicator species fed on.
The feeding habits of organisms allow for their classification based on what
they consume and therefore, the role this consumption plays in the
ecological integrity of a community (Moog, 1995). Organisms are generally
categorised into functional feeding-guilds based on the morphology of
mouthparts, feeding behaviour and food consumed (Cummins, 1973, 1974;
Cummins and Klug, 1979; Merritt and Cummins, 1984). Autecological
studies on the feeding behaviour of specific species are rarely available and
therefore there has been many false classification lists presented in the
77
literature. The following circumstances often prevent the unambiguous
classification of organisms into functional-feeding groups:
• Few species are obligate feeders on a specific food resource and many
use and select a wide range of items
• Certain organisms shift their diet during their life cycle and seasonally in
response to food availability
• Feeding classifications can be difficult to define as some organisms are
opportunistic feeders with unspecified nutritional requirements (Moog,
1995)
Cummins (1973) stated that the functional feeding group denotes a
hypothetical particle size range ingested or mode of feeding, not food type
or resource assimilation. Based on particle size ingested and mouthpart
morphology, the functional-feeding groups are important in that they allow
grouping of benthic communities into components. Functional feeding
groups could therefore be helpful when describing the mode of feeding or
food attainment and the role of the invertebrates in the processing of food
(Cummins and Klug, 1979; Wallace and Webster, 1996).
It is currently recognised however, that the functional feeding group does
not denote a specific, obligate trophic status for a given taxon and that the
function and/or trophic status can change between life stages of that taxon
(Cummins, 1988). It is difficult to classify an organism into a functional
feeding guild simply through evaluating its anatomy or morphology. The
role of the organism within the community also needs to be examined
closely in conjunction with detailed descriptions of its feeding preferences.
Ecologists are faced with the dilemma as to how to clarify the trophic
position (e.g. detritivore, herbivore, omnivore) and food web connections of
aquatic organisms. Food web investigations have traditionally been based
upon gut analyses or literature sources of food habits. The exact sources of
energy for animals are often not clear, especially for detritivores or
omnivores making interactions in food webs poorly understood (Paine,
1988; Mihuc and Toetz, 1994). It is therefore important to obtain
knowledge of the major components of the food web in an aquatic
78
ecosystem and how efficiently energy is transferred through the trophic
levels. The use of stable isotopes is also a powerful research tool to identify
interactions in food webs: e.g. detritus as an energy source in general
(Deegan et al., 1990), energy sources for animals in subalpine lakes (Rau,
1980; Estep and Vigg, 1985), the role of predaceous zooplankton in pelagic
food webs (Kling et al., 1992) and the use of allochthonous and
authochthonous food in streams (Rounick et al., 1982). The type of food
actually assimilated by the animal can be calculated using stable isotope
analyses but this can be very costly and time consuming.
A basic aspect of the structure and function of a freshwater ecosystem is its
material cycling and energy flow. A significant portion of such cycling and
flow involves the processing of various forms of organic matter by
freshwater invertebrates. This represents a basis for interest in aquatic
trophic relations including food intake, tissue assimilation and waste release
(Cummins, 1973). Studies carried out by Hynes (1970) found that the
diversity of the ingested food greatly exceeds the diversity of the aquatic
insects and that the majority of species appear to be generalists rather than
specialists. Reports into feeding habits have been subject to considerable
variation and require qualification with regard to habitat and species
preferences.
The functional feeding group concept widely used in aquatic ecology
(Cummins, 1973) is useful to specify how animals capture their food.
Analysis of functional feeding guilds provides an insight into the dynamic
ecological relations between construction, reconstruction and the
mineralisation processes. A functional feeding guild distribution
programme offers a method of indirect and empirical assessment of these
processes (Moog, 1995). A shift in the equilibrium state of a river-type,
from one governed by production to one dominated by decomposition
would indicate a disturbance. In turn, this disturbance would be reflected in
compositional changes in the feeding guild structure (Schweder, 1992;
Kohmann, et al., 1993; Moog 1993, 1994).
79
There is little information in relation to assimilation of the various resources
among lotic primary consumers. Some hypotheses have suggested that if
taxa eliminate most competitive interactions by partitioning space and/or
time, then generalist food utilisation would be expected (Schoener, 1974;
Townsend, 1989; Pahl-Wostl, 1993). Recent work by Mihuc and Minshall
(1995) and Mihuc (1997) has confirmed the argument put forward by
Cummins (1973), that many species of stream invertebrates are trophic
generalists capable of feeding and growing on a broad range of food types
and that the invertebrates also switch among resources in space and time.
Work carried out by Lamberti and Moore (1984) also suggest that most
aquatic insects are opportunistic feeders that consume a wide variety of food
items with seasonal and age-specific variation in feeding habits and diet.
Assimilation of food resources however, cannot be inferred from diets
unless the assumption is made that all of the material ingested by an
individual is assimilated, this assumption being unacceptable (Cummins,
1973; Warren, 1989; Martinez, 1993; Mihuc and Minshall, 1995). Often
assimilation efficiency values are inferred from limited published data,
where the same values for related taxa (Benke and Wallace, 1980; Smock
and Roeding, 1986), and sometimes for entire functional feeding groups
were found (Lugthart and Wallace, 1992). Based on current literature,
assimilation of both allochthonous (detrital fine and coarse particulate
organic matter; FPOM, CPOM) and autochthonous (periphyton, algae)
resources by lotic macroinvertebrates have been determined in relatively
few studies (Mihuc, 1997).
Environmental stress can cause disruptions in ecological integrity where the
primary effect is on autecological relations. Organisms respond to various
changes in the environment, which in principal are definable but it is
impractical to measure all but a few. There are species which react with
specific environmental factors, and produce characteristic responses that
alter the ecological function of a particular community (Moog, 1995).
Disturbance of an aquatic ecosystem (e.g. de- or afforestation, nutrient input
causing increased algal growth), can lead to changes in food availability
(Stout et al., 1993). The ability of an aquatic system to cope with such
disturbances is without doubt related to the feeding plasticity of the
80
organisms that inhabit it (Friberg and Jacobsen, 1994). For trophic
relationships and feeding plasticity to be understood, studies of feeding
preferences are essential, especially as invertebrates appear to prefer diets
most favourable to survival (Otto, 1974; Inversen, 1974; Kostalos and
Seymour, 1976).
Algal grazing is now acknowledged to be a very important link in the food
webs of streams in general (Lamberti and Moore, 1984; Hildrew, 1992).
Benthic algae form an important part of the biofilm that coats the upper
surfaces of substrate in running water (Madsen, 1972; Rounick and
Winterbourn, 1983; Lock et al., 1984). Other materials make up the mixed
assemblages on substrate surfaces, like coccoid bacteria, organic matter and
fungi. The living cells exude polysaccharide, which attaches them to the
substratum and forms a slimy matrix into which exoenzymes and exudates
are released (Lock, 1992). These activities enable interactions among the
cells, which may affect the growth and viability of populations making up
this complex community (Ledger and Hildrew, 1998).
It has been well documented that the rates of development and accrual of
algae on stones in streams are affected by many environmental variables;
among them are irradiance (light) (Ledger and Hildrew, 1998; Hill et al.,
1995), nutrient concentrations (Elwood et al., 1981; Stockner and Shortreed,
1976; Peterson et al., 1985), temperature (Darley, 1982), current velocity
(Peterson and Stevenson, 1992), substratum composition (Blinn et al., 1980;
Bott, 1983), disturbance (Grimm and Fisher, 1989; Peterson and Stevenson,
1992; Biggs and Thomsen, 1995) and the pH of the water (Maurice et al.,
1987). Deviations among these variables result in temporal and spatial
variations in the biomass and composition of biofilm, which in turn affect its
potential nutritional quality for invertebrate consumers (Ledger and
Hildrew, 1998). Grazing by macroinvertebrates has been shown to have an
overriding influence on periphyton in many situations. These studies
highlight the effects of grazing by decreasing biomass levels and
influencing or changing the taxonomic composition and community
structure of the periphyton (Hunter, 1980; Kelser, 1981; Lamberti and Resh,
1983; McAuliffe, 1984; Cattaneo and Kalff, 1986; Steinman et al., 1991;
81
Hill et al., 1992). Welch et al. (1992) found that lack of suitable habitat for
grazers was an important factor allowing the development of dense mats of
filamentous green algae, like Cladaphora, in some environments. Stream
herbivores can be as important as other factors such as light, in shaping
periphytic assemblages, although their mode of action in this respect is quite
different (Wellnitz and Ward, 1998). By eating or otherwise disturbing
periphytic algae, herbivores may change algal composition and physical
growth forms over others (Steinman, 1996). Mouthpart structure, ingestion
rates and foraging behaviour vary between taxa, and each of these traits may
influence the responses of periphytic standing crop, taxonomic composition
and algal physiognomy to grazing (Feminella and Resh, 1991; Lamberti et
al., 1987).
A close correspondence between food ingested and food assimilated would
be expected from evolutionary processes but the presence of a material in
the digestive tract does not prove nutritional importance. In other words
food habits may yield no information about assimilation. Significant
amounts of mineral sediment may be found in the guts of many periphyton
grazers and fine particle detritivores (Coffman, 1976; Coffman, et al., 1971;
Maciolek and Tunzi, 1968). The mineral substance is not nutritionally
significant but coatings of adsorbed organic material and associated bacteria
may be extremely important (Cummins, 1973). Furthermore, Mihuc, (1997)
notes however that these organic materials and associated bacteria are
important in establishing species food preferences and can be useful when
combined with assimilation efficiency estimates from the literature (but this
method does not actually measure assimilation directly).
The effect of food on regulating the life history characteristics of aquatic
insects has been investigated mostly in shredder and collector species (Otto,
1974; Ward and Cummins, 1979; Fuller and Mackay, 1981; Vannote and
Sweeney, 1985; Sweeney et al., 1986a, 1986b). Similar observations for
grazer-scrapers are scarce. Studies carried out by Hill and Knight (1987),
found that caddisfly larva with scraping mandibles were better at feeding on
inconspicuous algae than mayfly larva with collector-gatherer (grazer)
mouthparts. These collector-gatherers reduced the loose overlying layer
82
however, and they had a greater influence on community structure than the
scrapers.
Moog (1995) classifies aquatic organisms into functional feeding guilds
based on what they consume. He categorises the genus Ecdyonurus as a
grazer and detritus feeder (gathering collector) which has a preference for
epilithic algal tissues, biofilm and particulate organic matter (POM).
Detritus feeders (gathering collectors) tend to eat sedimented fine particulate
organic matter (FPOM). He implies that this genus may function as both a
grazer and a detritus feeder and this area clearly requires further
investigation.
In natural populations, gut contents of scraper-grazers usually contain
variable proportions of two main kinds of food: periphyton and fine
amorphous detritus (Chapman and Demory, 1963: Anderson and Cummins,
1979; Hawkins, 1985, Yule, 1986; Wallace and Gurtz, 1986). In general,
mouthpart morphology may largely determine the kinds of food to which
grazers have access (Gregory, 1983; Steinman et al., 1987); Karouna and
Fuller, 1992). There are a variety of different mouthpart morphologies
within the grazer community. The genus Ecdyonurus has brush-like labial
palps that engage in broad sweeping motions to harvest food from substrata,
which it then moves towards its mouth. It sweeps wide swaths of
periphyton from the substrata with its labial palps. It displays a clinging
mode of existence and possesses a dorsal-ventrally compressed head with
prognathous mouthparts (Arens, 1989). As with other larva from the
Heptageniidae family, Ecdyonurus venosus is a common grazer found in
most Irish rivers but there have been no studies undertaken in this country to
analyse its feeding regime, or to verify its assignment into this functional
feeding group. Other autecological data including species-specific
information about link relationships and energy transfer within the food
webs would also be required to accurately characterise the community
trophic relationships of this genus (Schoener, 1974; Schoenly et al., 1991;
Hildrew, 1992: Pahl-Wostl, 1993; Closs and Lake, 1994; Polis, 1994).
These investigations however were beyond the scope of this study, which
focuses entirely on feeding regimes alone.
83
3.2 Study outline
This study focuses on the feeding ecology of the mayfly larva, Ecdyonurus
venosus. One of the underlying hypotheses of this study is that as
eutrophication progresses, the periphyton species change thereby affecting
the food sources of Ecdyonurus. Specimens were obtained before and after
the addition of phosphorus to the river in order to assess any resultant
changes in feeding regime. In this study, a method permitting quantitative
examination of particulate gut contents for gut analysis of the mayfly larvae
Ecdyonurus venosus was adopted. This method allowed separation of
particles that make up the gut, into several distinct classes that are described
in detail below.
84
3.3 Materials and Methods
3.3.1 Site description: The Castlebar River
Details of the Castlebar River study site are given in Chapter 1.
3.3.2 Sampling design
The study was conducted at the split-stream experimental site in the
Castlebar River during the period of the nutrient manipulation. The river
was split using a 20m Perspex barrier separating each side uniformly. Prior
to the addition of the nutrient, each section of the split-stream was sampled
for macroinvertebrates. On 18th
July 2003 (pre-nutrient addition), five
Surber samples were taken at 4m intervals along each side of the divide.
From each Surber, specimens of the genus Ecdyonurus were carefully
collected and immediately preserved in 70% IMS (Industrial Methylated
Spirits). There was only one species of Ecdyonurus found in this river
during the study, namely Ecdyonurus venosus. Specimens were collected
again at intervals over a 7-week period post-nutrient addition. The
periphyton biomass was measured during the course of the experiment by
estimating the chlorophyll a concentrations (µg/cm2) from stone scrapings.
The mean benthic chlorophyll a concentrations in the treated section
showed statistically significant increases towards the end of the study in
comparison with the control section (see Chapter 2). The concentrations
seemed to increase particularly in the last three weeks of the study (Fig.
3.1). It was hypothesised that there may have been a change in the feeding
regime of Ecdyonurus caused by the increase in the periphyton biomass on
the stones in the treated section, found in the latter stages of the study.
Specimens were chosen from the treated section for gut content analysis
from samples obtained on 12th
September 2003 when the chlorophyll a
concentrations were elevated compared to the control section (Fig. 3.1).
85
Fig. 3.1 Mean chlorophyll a concentrations (µg/cm2) in the treated section (manipulated)
vs the control section. Bars are means ± standard deviation: n = 5.
3.3.3 Gut content analysis
3.3.3.1 Cold acid hydrogen peroxide method
Larvae of Ecdyonurus venosus were sampled from a number of high status and impacted
rivers in March 2003 (Table 3.7 and 3.8). The cold acid hydrogen peroxide method was the
first and most simplistic technique used in an attempt to isolate and identify the material
consumed by this species. This study was carried out to provide species-level information
on the diatoms consumed by this larva and did not provide information on the overall
particulate material eaten. The DAPI method described below was used to categorise the
material consumed in more detail.
The cold acid hydrogen peroxide method was used to isolate the benthic diatoms in the gut
contents. The gut was dissected from each animal and exposed to cold 30% hydrogen
peroxide with potassium permanganate to aid oxidation and digestion. These samples were
then washed several times with distilled or deionised water to removed any chemicals,
centrifuged and allowed to sediment overnight. The supernatant was decanted the
following morning. A known volume of material was mounted onto a microscope slide,
covered with a cover slip and heated gently on a hotplate to fix. A total of 15-20 fields
were observed using oil immersion (10x100 magnification). An attached camera was used
to take photographs. The gut contents of eight larvae were oxidised and seven were left
untreated and intact for direct observation under the microscope. These were used for
comparative purposes, in the event that oxidation led to loss of sample.
0
50
100
150
200
250
21/0
7/03
28/0
7/03
04/0
8/03
11/0
8/03
18/0
8/03
25/0
8/03
01/0
9/03
08/0
9/03
15/0
9/03
Date
Ch
lorop
hyll
a u
g/c
m2
control section treated section
86
3.3.3.2 DAPI method
Gut content particles ingested by the mayfly larvae were examined using a
fluorochromatic stain 4’6 diamidino-2-phenylindole (DAPI) with nuclear
DNA-specific binding properties; DNA-bound DAPI fluoresces blue when
exposed to light at 365nm and viewed at high magnification (Wittekind,
1972). This stain was used in conjunction with epifluorescence microscopy
(Walker et al., 1988) and with light microscopy to identify the particles
consumed by the mayfly larvae. The DAPI stain allows for the enumeration
of very small bacteria, below the resolution of a light microscope (<1.0µm)
due to the fluorescent glow of the bacterial DNA. This stain has been
successfully used to quantify bacterial counts in waters rich in organic
material (Kondratieff and Simmons, 1985; Rassoulzadegan and Sheldon,
1986).
Five Ecdyonurus specimens of various sizes were selected from samples
obtained from each section of the divide on 18th
July 2003 and 12th
September 2003. In total 20 individual Ecdyonurus venosus specimens
were examined for gut analysis in this study. Ledger and Hildrew, (2000),
in their study of the gut contents of stoneflies and chironomids,
recommended removing food particles from the foregut, where food tends to
be more intact than in the mid or hindgut. This area was difficult to
pinpoint when dissecting the guts of the mayfly larva. As no descriptions
could be found in the literature describing exactly where the fore-, mid- or
hindguts were located in the digestive tract, the gut was visually split into
three equal sections and assigned fore- mid- and hindgut accordingly.
Occasionally there was no food in the foregut so the mid- and hindgut had
to be taken as a whole, while there were other times when food was only
present in the hindgut. Each area removed for examination was carefully
noted. The animals were measured to the nearest millimetre from the top of
the labium to the tip of the last abdominal segment. The gut of each larva
was removed under a dissecting stereomicroscope using a fine forceps and
the contents from the gut were separated from the peritrophic membrane
with a scalpel and fine forceps into a drop of distilled water in a petri dish.
87
The material was pipetted from the drop into an Eppendorf tube with 2mls
of distilled water and vortexed for 20-30seconds. DAPI solution was added
(0.2mls) to each vial and held in the dark for 20-25mins. To provide a dark
background for improved visibility of the fluorescing particles, 25mm black
Isopore (0.2µm) filters were used. For even distribution, these filters were
soaked in distilled water and placed on top of a damp backing filter
(Millipore, 25mm diameter, 0.45µm pores size) in a small filter holder
(Swinnex W/O filter). After staining, the sample was shaken and filtered
through the Swinnex filtration system (Hobbie et al., 1977; Porter and Feig,
1980). A BX60 Olympus microscope with an Olympus U-RFL-T
epifluorescence burner was used to view the fluorescent stained microscope
slides. Photomicrography was done using an attached Olympus camera.
The damp filters were removed and placed onto microscope slides
previously smeared with immersion oil. A drop of immersion oil was added
to the filter and a coverslip was then placed on top. The slides were stored
horizontally in the dark at 4°C. All particles in 15 fields of view were
counted and assigned to one of several groups.
Particles found in the gut of Ecdyonurus venosus were assigned to the
following categories: (1) filamentous green algae (2) coccoid green algae
(3) diatoms (4) biofilm matrix (5) algal agglomerates (6) inorganic debris
(7) plant detritus (8) coccoid bacteria (9) peritrophic membrane and (10)
unidentified particles.
When the DAPI stain is exposed to UV-light, bacteria and algal nuclei
fluoresce bright blue while protozoans appear light blue with a distinct
nucleus. Chlorophyll in algae autofluoresces a dull red/orange colour and
organic plant detritus fluoresces dull yellow (Coleman, 1980; Porter and
Feig, 1980; Rassoulzadegan and Sheldon, 1986). Illuminating the sample
further with white light enabled clumps of algal agglomerates to be
distinguished from the biofilm matrix. Algal agglomerates autofluroesce
red under white light highlighting the algal chloroplasts. The bacterium
attached to the matrix in biofilm does not autofluoresce under white light
thereby disappearing when switched between channels (from UV-light to
red). This was useful, as sometimes it was difficult to distinguish between
88
these categories as bacteria and algal nuclei fluoresce bright blue causing
confusion on occasion. The structure of biofilm matrix consists of a thin
film of brown matrix material with a porous or granular appearance, which
generally contains algal matter, embedded with bacteria (Ledger and
Hildrew, 2000). Algae were assigned to an ‘algal agglomerate’ category
when the algae appeared as a mass of algal chloroplasts as opposed to a
filamentous green algae or coccoid green algal particle. As Ledger and
Hildrew (2000) found, some particles in this study could not be identified so
they were assigned to an ‘unidentified particles’ category along with detrital
material of unknown origin. On occasions, peritrophic membrane was
found in the sample and was estimated as part of the material examined but
not reported in the results. Therefore, peritrophic membrane was discounted
when expressing percentage composition results.
The technique is a relatively time-consuming one. Some preliminary
samples were analysed in order to develop the dissection technique and
perfect the staining process. Once the techniques were developed the
dissection, counting and photographing of the 20 Ecdyonurus gut contents
took 5 weeks.
89
3.3.4 Analysis of Benthic stone scrapings
Stone scrapings were obtained from submerged cobbles prior to the addition
of nutrient and again at intervals during the manipulation experiment.
These were taken to quantify the composition of algal communities growing
on the substrate and for taxonomic comparisons with those found in the
Ecdyonurus gut analysis. Qualitative stone scrapings were taken from 5
stones at various intervals along each side of the divide and they were
pooled into a 100ml-perspex glass bottle. The samples were preserved
immediately by adding 0.7ml of Lugols iodine and stored in the dark for
later analysis. Two samples from scrapings obtained on 21st July 2003 (pre-
nutrient addition) and two samples from scrapings taken on 9th
September
2003 (post-nutrient addition) were examined in this part of the investigation.
These dates were chosen as they coincided with larval sampling periods.
Ecdyonurus venosus has been classified as a grazer and thus it would be
expected to move in broad sweeping motions to harvest food from the
substrate by brushing with its labial palps (Wellnitz and Ward, 1998). They
collect various particulate types as they feed and it was hypothesised that
there may be a similarity between the material taken from the stone
scrapings and the food ingested by the larvae. In order to quantify the
composition of the stone scrapings, they were analysed using the same
epifluorescent microscopy technique as was used to examine the gut
contents. Lugols iodine has historically been used to preserve
phytoplankton but its use in combination with the DAPI stain to identify
particulate material as described above was not previously documented.
Stone scrapings obtained from each side of the experimental divide on
21/07/03 and 09/09/03 were chosen as they coincided with larval sampling
periods. The Lugols acts as a fixative-preservative stain, that makes cells
dark brown but can mask chlorophyll fluorescence (Gifford and Caron,
2000). Thus it was not known how effectively the DAPI stain would work
in conjunction with the Lugols.
For quantitative algal taxa observations, a 1ml aliquot was taken from each
preserved suspension, placed in a 10ml settling chamber, filled to the top
90
with distilled water and left to settle for 24 hours. The algal taxa were
identified and enumerated along transects across the width and length of the
chamber until a total of 15 fields were viewed per sample.
91
3.4 Results
3.4.1 Gut content analysis
The gut contents of Ecdyonurus venosus specimens were analysed for
samples taken on five separate dates from the Castlebar River. Details of
the sampling dates and the number of specimens dissected are outlined in
Table 3.1.
A chi-squared analysis was carried out on different treatment combinations
of the gut contents of Ecdyonurus venosus and the stone scrapings taken
from the Castlebar River during the summer of 2003 to establish if there
were any similarities between the gut contents and the food material scraped
from the stone surfaces (Table 3.11). The results are outlined in section
3.4.3.
Table 3.1 Details of the sampling dates and the number of specimens
examined for the gut content investigations in the Castlebar River in 2003.
Sampling date Number of specimens Number of specimens
Split stream nutrient addition
comparison
Treated Side Control Side
18th
July 2003 (pre-treatment) 5 5
12th
Sep 2003 (post treatment) 5 5
Additional samples (see Table 3.4 for
details) Number of specimens
29th
July 2003 2
7th
August 2003 1
19th
August 2003 3
Specimens were examined and food items were assigned to one of ten food
categories previously outlined. Coccoid bacteria were relatively common in
the guts but these were not included in the percentage composition results of
any of the specimens below as they generally represented less than 1% of
the total particle area in the majority of the invertebrate gut contents
investigated.
Details of the length of the individual specimens examined and the area of
the digestive tract dissected are outlined in Tables 3.2, 3.3 and 3.4.
92
Table 3.2 Length of the Ecdyonurus venosus specimens (mm) examined
and details of the area of the digestive tract dissected during the gut analysis
investigations pre-nutrient addition (18th
July 2003).
Length specimen
(mm)
Area of digestive
tract dissected
Control section 10.0 Fore-, mid- and hindgut
9.5 Fore-, mid- and hindgut
8.0 Fore-, mid- and hindgut
8.2 Mid- and hindgut
8.0 Mid- and hindgut
Treated section 9.4 Fore-, mid- and hindgut
11.5 Mid- and hindgut
10.0 Fore-, mid and hindgut
10.0 Fore- and midgut
9.0 Mid- and hindgut
Table 3.3 Length of the Ecdyonurus venosus specimens (mm) examined
and details of the area of the digestive tract dissected during the gut analysis
investigations post-nutrient addition (12th
September 2003).
Length specimen
(mm)
Area of digestive
tract dissected
Control section 8.5 Mid- and hindgut
9.0 Midgut
8.5 Mid- and hindgut
7.4 Mid- and hindgut
8.0 Midgut
Treated section 8.0 Foregut
8.0 Mid- and hindgut
7.0 Fore- and midgut
7.3 Midgut
7.5 Mid- and hindgut
Table 3.4 Additional Samples: length of the Ecdyonurus venosus specimens
(mm) examined and details of the area of the digestive tract dissected during
the gut analysis studies.
Date Length specimen
(mm)
Area of digestive tract
dissected
Control section 29/07/03 10.2 Fore- and midgut
Treated section 29/07/03 11.4 Fore- and midgut
07/08/03 11.7 Fore- and hindgut
19/08/03 10.9 Fore-, mid- and hindgut
19/08/03 11.8 Fore- and midgut
19/08/03 11.1 Fore- and midgut
93
The mean percentage composition of the gut contents in the ‘pre-nutrient
addition’ specimens (18th
July 2003) and in the ‘post-nutrient addition’
specimens (12th
September 2003) are outlined in Table 3.5.
Table 3.5 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the Castlebar River on 18th
July 2003 and 12th
September 2003.
Pre-nutrient
addition 18th
July 2003
Pre-nutrient
addition 18th July
2003
Post-nutrient
addition 12th
September 2003
Post-nutrient
addition 12th
September 2003
Food category Treated section Control section Treated section Control section Diatoms - - 2% 7%
Desmids 1% 4% - -
Filamentous green algae 8% 4% 5% 4%
Coccoid green algae 11% 15% 8% 7%
Biofilm matrix 4% 5% 4% 2%
Inorganic debris 24% 16% 16% 30%
Unidentified particles 3% 4% 15% 7%
Plant detritus 45% 41% 37% 26%
Algal agglomerates 4% 11% 13% 17%
3.4.1.1 Gut content results ‘pre-nutrient addition’ (18th
July 2003)
The mean percentage composition of gut contents in the treated section of
the Castlebar River on 18th
July 2003 (pre-nutrient addition) is shown in
Fig. 3.2. The results for the control section of the Castlebar River on the
18th
July 2003 are shown in Fig. 3.3. Photomicrographs of the DAPI stained
particles found in the gut contents of Ecdyonurus venosus ‘pre-nutrient
addition’ (18/07/03) are shown in Fig. 3.4A-F. Coccoid bacteria (not shown
in graphs) represented 0.4% and 2.7% of the total particulate area of the gut
contents of the specimens examined in the treated and control sections pre-
nutrient addition, respectively.
94
Fig. 3.2 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the treated section of the Castlebar River pre-nutrient addition.
Bars are + standard deviation: n = 5.
Fig. 3.3 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the control section of the Castlebar River pre-nutrient addition.
Bars are + standard deviation: n = 5.
The gut contents of the specimens examined prior to the addition of nutrient,
consisted of algae (desmids, coccoid green algae, filamentous green algae
and algal agglomerates), quite a large proportion of plant detritus, biofilm
matrix, inorganic debris and unidentified particles. The mean composition
of ingested material in the larvae examined from both sides of the split-
stream prior to nutrient addition, was proportionately relatively similar
(Table 3.5) with no significant differences observed between the food items
(Table 3.11; p=0.3). Plant detritus was the dominant material in the gut of
these mayflies, comprising nearly half of the total particulate area. It
0
20
40
60
80
100
Des
mid
s
Fil
amen
tou
s
gre
en a
lgae
Co
cco
id
gre
en a
lgae
Bio
film
mat
rix
Ino
rgan
ic
deb
ris
Un
iden
tifi
ed
par
ticl
es
Pla
nt
det
ritu
s
Alg
al
agg
lom
erat
es
% C
om
po
siti
on
0
20
40
60
80
100
Dia
tom
s
Fil
amen
tou
s
gre
en a
lgae
Co
cco
id
gre
en a
lgae
Bio
film
mat
rix
Ino
rgan
ic
deb
ris
Un
iden
tifi
ed
par
ticl
es
Pla
nt
det
ritu
s
Alg
al
agg
lom
erat
es
% C
om
po
siti
on
95
accounted for 41% in the control section compared to 45% in the treated
side of the experiment. Filamentous green algae were found to represent
4% (control) and 8% (treated) of the larvae guts, coccoid green algae
represented 11% (treated) and 15% (control) and algal agglomerates
accounted for 4% (treated) and 11% (control) on either side of the
experimental divide. Desmids were represented only by Closterium spp.,
which was found in low numbers accounting for 1% (treated) and 4%
(control) of the gut contents of Ecdyonurus venosus during this sampling
period. Biofilm matrix was present in low amounts comprising 4% (treated)
and 5% (control) of the guts examined, as were unidentified particles
ranging from 3% (treated) and 4% (control). Inorganic particles were
present, representing 24% (treated) and 16% (control) of the total particle
area in the guts examined between each section of the split-stream.
96
Fig. 3.4A 400x A possible nematode spp.
Specimen unravelling in the gut of an 11.5mm
specimen of Ecdyonurus venosus during the
digestion process. This specimen was taken
from the treated section of the split-stream
experiment on 18th
July 2003.
Fig. 3.4B 200x Closterium spp. (desmid)
found in the gut of an 11.5mm
Ecdyonurus venosus specimen in
the treated section of the split-stream
experiment on 18th
July 2003.
Fig. 3.4C 400x Branched filament of algae
with nuclei brightly fluorescing in the centre
of each cell. Found in the gut of a 12mm
E.venosus specimen in the treated section of
the split-stream experiment on 18th
July 2003.
Fig. 3.4D 200x Detritus and inorganic
particles found in the gut of a 9.5mm
specimen of E.venosus. A section of
peritrophic membrane of the animal is
fluorescing bright blue. The animal was
sampled from the control section of the
experiment on 18th
July 2003.
Fig. 3.4E 400x Algal agglomerate in the gut
of a 7.4mm specimen of E.venosus taken from
the control section of the split-stream
experiment in the Castlebar River on 18th
July
2003.
Fig. 3.4F 200x Cocconeis spp. (egg shaped)
with another partially digested of same to
the foreground. Plant detritus and inorganic
particles are scattered throughout the
sample. Found in the same gut as in Fig 3.6E
97
3.4.1.2 Gut content results ‘post-nutrient addition’ (12th
September 2003)
The mean percentage composition of the gut contents from samples
obtained from the treated section of the split-stream experiment, on 12th
September 2003 (post-nutrient addition) is shown in Fig. 3.5. Results for
the control section are given in Fig. 3.6. Photomicrographs of the DAPI
stained particles found in the gut contents of specimens post-nutrient
addition (12/09/03) are shown Fig 3.7A-F. Coccoid bacteria (not shown in
graphs) represented 0.1% and 0.6% of the total particulate area of the gut
contents of the specimens examined in the treated and control sections post-
nutrient addition, respectively.
Fig. 3.5 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the treated section of the Castlebar River post-nutrient addition.
Bars are + standard deviation: n = 5.
Fig. 3.6 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the control section of the Castlebar River post-nutrient addition.
Bars are + standard deviation: n = 5.
0
20
40
60
80
100
Dia
tom
s
Fil
amen
tou
s
gre
en a
lgae
Co
cco
id
gre
en a
lgae
Bio
film
mat
rix
Ino
rgan
ic
deb
ris
Un
iden
tifi
ed
par
ticl
es
Pla
nt
det
ritu
s
Alg
al
agg
lom
erat
es
% C
om
po
siti
on
0
20
40
60
80
100
Dia
tom
s
Fil
amen
tou
s
gre
en a
lgae
Co
cco
id
gre
en a
lgae
Bio
film
mat
rix
Ino
rgan
ic
deb
ris
Un
iden
tifi
ed
par
ticl
es
Pla
nt
det
ritu
s
Alg
al
agg
lom
erat
es
% C
om
po
siti
on
98
Fig. 3.7A 400x Navicula spp. found in the gut
of a 7mm E.venosus in the treated section of
the split-stream experiment on 12th
September
2003.
Fig. 3.7B 400x Algal agglomerate (green
mass to left of photo) with a string of
coccoid bacteria and plant detritus. Found in
a 7.3mm E.venosus specimen in the treated
section of the Castlebar River on 12th
September 2003.
Fig. 3.7C 200x Branched filament of algae
with plant detritus autofluorescing yellow.
Found in the gut of an 8mm E.venosus
specimen in the treated section of the
Castlebar on 12th
September 2003.
Fig. 3.7D 200x Closterium spp. (Desmid –
150µ) found in the gut of an 8mm specimen
of E.venosus in the treated section of the
Castlebar River on 12th
September 2003.
Fig. 3.7E 200x Biofilm matrix in the gut of an
8.5mm specimen of E.venosus in the control
section of the Castlebar River on 12th
September 2003.
Fig. 3.7F 200x Filament of algae found in
the gut of an 11mm specimen of E.venosus
in the control section of the split-stream
experiment. Plant detritus (dull yellow
coloration) is also evident.
99
In this case the larval gut contents consisted primarily of algae (diatoms,
coccoid green algae, filamentous green algae and algal agglomerates). They
also contained quite a large proportion of plant detritus and inorganic debris.
Unidentified particles and biofilm matrix made up a proportion of the larva
guts. As with the specimens investigated on 12th
July (pre-nutrient
addition), the larvae studied on this occasion contained a good proportion of
plant detritus (26% and 37%) in each side of the experimental divide.
Navicula spp., was found in specimens sampled from the treated section
representing 2% of total particulate content. This species accounted for 7%
of the gut content of animals taken from the control section of the split-
stream on the same day (12th
September). Filamentous green algae
represented 5% (treated) and 4% (control) of the gut contents in both study
sections. Coccoid green algae and algal agglomerates represented 8%
(treated) and 13% (treated) and 7% (control) and 17% (control) respectively.
Biofilm matrix makes up for 2% (control) and 4% (treated) of the
particulate material and inorganic debris accounts for 30% (control) and
16% (treated) of the contents of guts analysed on the treated and control
sections of the experimental divide respectively (Table 3.5).
Significant differences in the gut contents were found between the treated
and the control sections post-nutrient addition (Table 3.11 Treatment
combination 6 – p<0.001). The main difference in food composition
between summer and autumn sampling, was the absence of diatoms in the
guts examined in July, which were present in the larval stomachs in
September. Greater numbers of diatoms are present in rivers during the
spring and autumn seasons, which may account for their presence in
September. There were also desmids found on occasion in the specimens
examined in July only. Other specimens from the Castlebar River studied
during the investigation did not contain diatoms or desmids. These latter
did however contain filamentous green algae, coccoid green algae, biofilm
matrix and inorganic debris. Plant detritus also dominated the diet of these
larvae. The most common diatom species found in the ingested material
was Navicula spp. No other diatom was found in its gut contents, which
suggests that Navicula spp. was the main diatom species consumed during
this period. This diatom may also be more tolerant to the digestive
100
processes in the gut of these larvae compared with other taxa. The
significant difference found in the gut contents between the treated and the
control sections post-nutrient addition appear to be attributed to the natural
variation in the food items available to Ecdyonurus during a particular
feeding episode.
3.4.1.3 Gut content results from additional samples investigated
The gut contents of the larvae were also analysed for three other sampling
dates. These were carried out before the main left/right (control v treated
sections) comparison and formed a preliminary examination to verify that
the technique worked but it was felt worthwhile to report on these results as
well. Two guts were examined from samples obtained on 29th
July, one
specimen was investigated from samples taken on 7th
August and three from
samples taken on 19th
August 2003. With the exception of the larva
analysed on 29th July, which was taken from the control section of the
experimental divide, all other specimens were taken from the treated
section.
The percentage breakdown of the gut contents for these specimens is
outlined in Table 3.6. The food contents of the guts of Ecdyonurus venosus
examined are also shown in Figs. 3.8 to 3.11. Photomicrographs illustrating
examples of the DAPI stained particles, found in the gut contents of the
Ecdyonurus venosus specimens investigated on 29/07/03, are shown in Fig.
3.12A-F. More particulate material found in the gut contents of mayflies
examined on 07/08/02 and 19/08/03, are presented in Fig. 3.13A-F.
In addition to the items in the Table 3.6 and Figs. 3.7 to 3.10, coccoid
bacteria represented 0.3% of the total particulate area of the gut contents of
the specimens examined on 29/07/03. Coccoid bacteria represented 0.1%
and 0.2% of the total particulate area of the gut contents of the specimens
examined on 07/08/03 and 19/08/03 respectively.
101
Table 3.6 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the additional samples investigated in the Castlebar River during
the study.
29/07/03 29/07/03 07/08/03 19/08/03
N=1 N=1 N=1 N=3
Food category Treated
section
Control
section
Treated
section
Control
section Diatoms - - - -
Desmids - - - -
Filamentous green algae 13% 13% 1% 16%
Coccoid green algae 32% 9% 5% 16%
Biofilm matrix 0% 23% 5% 10%
Inorganic debris 16% 23% 20% 20%
Unidentified particles 23% 11% 5% 8%
Plant detritus 16% 21% 64% 30%
Algal Agglomerates - - - -
Fig. 3.8 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the control section of the Castlebar River on 29/07/03.
Fig. 3.9 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the treated section of the Castlebar River on 29/07/03.
0
20
40
60
80
100
Fil
amen
tou
s
gre
en a
lgae
Co
cco
id
gre
en a
lgae
Bio
film
mat
rix
Ino
rgan
ic
deb
ris
Un
iden
tifi
ed
par
ticl
es
Pla
nt
det
ritu
s
% C
om
po
siti
on
020406080
100
Fil
amen
tou
s
gre
en a
lgae
Co
cco
id
gre
en a
lgae
Ino
rgan
ic
deb
ris
Un
iden
tifi
ed
par
ticl
es
Pla
nt
det
ritu
s
% C
om
po
siti
on
102
Fig. 3.10 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the treated section of the Castlebar River on 07/08/03.
Fig. 3.11 Mean percentage composition of the gut contents of Ecdyonurus
venosus in the treated section of the Castlebar River on 19/08/03. Bars are +
standard deviation: n = 3.
020406080
100
Fil
amen
tou
s
gre
en a
lgae
Co
cco
id
gre
en a
lgae
Bio
film
mat
rix
Ino
rgan
ic
deb
ris
Un
iden
tifi
ed
par
ticl
es
Pla
nt
det
ritu
s% C
om
po
siti
on
0
20
40
60
80
100
Fil
amen
tou
s
gre
en a
lgae
Co
cco
id
gre
en a
lgae
Bio
film
mat
rix
Ino
rgan
ic
deb
ris
Un
iden
tifi
ed
par
ticl
es
Pla
nt
det
ritu
s
% C
om
po
siti
on
103
Fig. 3.12A 100x Ecdyonurus spp. gill with
distinct filament found in the gut of a 10.2mm
E.venosus specimen in the control section of
the split-stream experiment on 29th
July 2003.
Fig. 3.12B 400x Femur of early invertebrate
instar with agglomerate of detritus in the gut
of a 10.2mm E.venosus specimen in the
control section of the split-stream
experiment on 29th
July 2003.
Fig. 3.12C 400x Peritrophic membrane of the
gut of a 10.9mm E.venosus specimen
fluorescing taken from the treated section of
the split-stream experiment on 29th
July 2003.
Fig. 3.12D 400x Filament of algae in spiral
formation with nematode spp. in the gut of
E.venosus in the treated section of the
Castlebar River on 29th
July 2003.
Fig. 3.12E 400x Coccoid green algal mass
showing the chloroplasts and nuclei
fluorescing under UV-light found in the gut of
a 10mm E.venosus specimen in the treated
section of the split-stream experiment.
Fig. 3.12F 400x Coccoid green algae mass
as in Fig. 2.20E but viewed under red light .
Notice how the bright blue fluorescence of
the chloroplasts and nuclei fade under this
light. The algae cell walls are highlighted.
104
Fig. 3.13A 400x Coccoid green algae
fluorescing bright blue in the gut of an
11.4mm E.venosus specimen in the control
section of the split-stream experiment on 7th
August 2003.
Fig. 3.13B 400x Coccoid green algae
embedded in algal mass with a nematode sp.
entwined in the agglomerate. Found in the
gut of an 11.4mm E.venosus specimen in the
control section of the Castlebar River on 7th
August 2003.
Fig. 3.13C 400x Algal agglomerate found in
the gut of a 10.5mm E.venosus specimen in
the treated section of the Castlebar River on
7th
August 2003.
Fig. 3.13D 400x Filament of algae showing
nuclei and chloroplasts fluorescing bright
blue. Found in the gut of a 10mm E.venosus
specimen in the treated section of the
Castlebar River on 19th
August 2003.
Fig. 3.13E 400x Coccoid bacteria fluorescing
bright blue under UV-light in the gut of an
11mm E.venosus specimen taken from the
control section of the split-stream experiment
on 19th
July 2003.
Fig. 3.13F 400x Photomicrograph taken
from the gut of a 9mm E.venosus specimen
in the control section of the Castlebar River
on 19th
July 2003. Plant detritus is clearly
autofluorescing.
105
The gut contents of the larvae investigated on these occasions (Table 3.6)
were similar to those studied on the 18/07/03 and 12/09/03 (Table 3.5). In
comparison to those investigated pre- and post-nutrient addition, all
specimens contained more filamentous algae with the exception of the
sample examined on 07/08/03. There were no diatoms and desmids found
in gut contents of the larvae examined on these other visits. Biofilm matrix,
plant detritus and coccoid green algae made up a large proportion of the gut
contents of these mayflies. The larvae ingested reasonable quantities of
inorganic debris (mineral particles) and unidentified particles (Table 3.6).
There were no algal agglomerates found in the ingested material of any of
the larvae studied. The gut contents of one specimen showed some quite
unexpected inclusions (Fig. 3.12A and B) containing the gill of an early
instar of Ecdyonurus with the filament intact, along with the femur of an
invertebrate instar with an agglomerate of detritus. This suggests that
Ecdyonurus is relatively unselective in its feeding pattern as the animal
appeared to ingest these animal remains together with its regular food items
while it moved across the surface of the river substrate during a feeding
episode.
The examination of these specimens provided preliminary results that
contributed to knowledge of the diet of Ecdyonurus venosus. These results
successfully validated the technique in isolating and identifying the material
in the gut contents. This method was then employed on the larvae examined
pre- and post-nutrient addition, which supplied the bulk of the information
on gut content analyses in this investigation.
106
3.4.1.4 Results of gut analysis of Ecdyonurus venosus during spring
2003 investigation
As part of the investigation into the feeding regime of Ecdyonurus venosus,
various other rivers were also sampled to obtain specimens for the gut
analysis studies. Specimens of this mayfly species were collected from five
high status rivers and five impacted rivers during March 2003. This
investigation was carried out to provide information on the types of benthic
diatoms consumed by these larvae. As diatoms are abundant in spring, it
was decided to examine the gut contents of specimens sampled during this
period. Intact samples were only present in three of the high status sites and
two of the impacted sites investigated (N=5). The specimens were
preserved in 70% IMS in the field and retained for later analysis. As
mentioned in the Materials and Methods section, some of the gut contents
were oxidised after dissection with 30% hydrogen peroxide, while the
remainders were examined without treating. This was carried out for
comparative purposes. The gut contents of each specimen was dissected
and viewed under a compound light microscope without using any staining
materials. Photomicrographs of the material found in these samples are
illustrated in Fig. 3.14A-I. A list of algal taxa that were found in the gut
contents of four Ecdyonurus venosus specimens taken from three high status
sites in March 2003 are presented in Table 3.7. A list of algal taxa that were
found in the gut contents of three specimens taken from two impacted sites
are presented in Table 3.8.
107
Fig. 3.14A 400x The gut contents of a 10mm
specimen of E.venosus sampled from the
Dunneill River in March 2003 showing an
abundance of diatoms.
Fig. 3.14B 400x Gomphonema trancatum
(Bacillariophyceae) found in the gut of a
10mm E.venosus specimen in the Dunneill
River in March 2003.
Fig. 3.14C 400x Achnanthes lancelota
(Bacillariophyceae) found in the gut of a 14mm
specimen of E.venosus in the Mullaghanoe
River in March 2003.
Fig. 3.14D 400x Rhoicosphenia abbreviata
(Bacillariophyceae) in broken girdle view in
the gut of a 14mm E.venosus specimen in
the Mullaghanoe River during March 2003.
Fig. 3.14E 400x Filament of green algae found
in the gut of a 14mm E.venosus specimen in the
Castlebar River in March 2003.
Fig. 3.14F 400x Gomphonema parvulum
(Bacillariophyceae) in the gut of a 10mm
E.venosus in the Dunneill River during
March 2003.
108
Fig. 3.14G. 400x Navicula lancelota
(Bacillariophyceae) found in the gut of a
14mm specimen of E.venosus the Castlebar
River in March 2003.
Fig. 3.14H. 400x Eutonia spp.
(Bacillariophyceae) found in the gut of a
15.6mm specimen E.venosus in Callow
Loughs Stream River during March 2003.
Fig. 3.14I. 400x Cocconeis placentula
(Bacillariophyceae) found in the gut of a
14mm specimen of E.venosus the Castlebar
River in March 2003.
109
Table 3.7 Algal taxa found in the gut contents of four specimens of
Ecdyonurus venosus, in three high status rivers during March 2003.
River Body length
(mm)
Algal taxa
Castlebar
Non-oxidised sample
14 Achnanthidium minutissimum *
Achnanthes lanceolata
Cocconeis placentula
Cymbella minuta
Fragilaria capucina
Fragilaria capucina var. vaucheriae
Gomphonema clavatum
Gomphonema parvulum
Meridion circulare
Navicula gregaria
Navicula lanceolata
Nitzschia dissipata
Reimeria sinuata
Rhoicosphenia abbreviata
Green filament–not long enough to
identify
Dunneill
Non-oxidised sample
13 Achnanthidium minutissimum
Diatoma moniliformis
Gomphonema olivaceum
Gomphonema parvulum
Navicula lanceolata
Reimeria sinuata
Dunneill
Non-oxidised sample
10 Achnanthidium minutissimum*
Achnanthes lanceolata
Cocconeis placentula
Cymbella minuta
Cymbella silesiaca
Diatoma moniliformis
Gomphonema parvulum
Gomphonema truncatum
Navicula lanceolata
Nitzschia dissipata
Reimeria sinuata
Achnanthes spp.
Callow Loughs Stream
Oxidised sample
15.6 Achnanthes biasolettiana
Achnanthidium minutissimum
Cocconeis placentula
Eunotia spp.
Fragilaria capucina var. vaucheriae
Gomphonema olivaceum
Meridion circulare
Navicula lanceolata
Nitzschia dissipata
Synedra ulna
*Achnanthidium minutissimum was the dominant algal taxa found in the gut contents of
these specimens of Ecdyonurus venosus.
110
Table 3.8 Algal taxa found in the gut contents of three specimens of
Ecdyonurus venosus in two-impacted rivers during March 2003.
River Body length
(mm)
Algal taxa
Robe
Non-oxidised sample
8.5 Reimeria sinuata*
Achnanthes lanceolata
Achnanthidium minutissimum
Amphora pediculus
Cocconeis placentula
Gomphonema olivaceum
Gomphonema parvulum
Navicula tripunctata
Achnanthes spp.
Mullaghanoe
Oxidised sample
15.6 Achnanthes lanceolata
Achnanthidium minutissimum
Cocconeis placentula Fragilaria capucina var. vaucheriae
Gomphonema angustum
Gomphonema micropus
Gomphonema parvulum
Meridion circulare
Navicula gregaria
Navicula lanceolata
Surirella brebissonii
Mullaghanoe
Non-oxidised sample
14 Achnanthes lanceolata
Achnanthidium minutissimum
Cocconeis placentula
Fragilaria capucina var. vaucheriae
Gomphonema angustum
Gomphonema micropus
Gomphonema parvulum
Meridion circulare
Navicula lanceolata
Nitzschia dissipata
Reimeria sinuata
Rhoicosphenia abbreviata
* Reimeria sinuata was the dominant algal taxa found in the gut contents of this specimen
of Ecdyonurus venosus.
111
The specimens examined from the high status and impacted rivers yielded
some interesting findings, showing an abundance of diatoms in the gut
contents (Tables 3.7 and 3.8) which corresponds with increased diatom
growth during the spring season. The ingested food of the invertebrates that
were not subject to oxidation tended to contain more diatoms than the
treated ones, suggesting loss of algal taxa when using the oxidation
technique. Only three of the eight oxidised samples examined contained
diatoms, which were present in small numbers. Four out of the seven
untreated samples contained diatoms and were particularly abundant in the
Castlebar and Dunneill Rivers.
The dominant algal taxon found in the stomachs of the animals in the
Dunneill River, the Castlebar River and Callow Loughs Stream (high status
rivers) was Achnanthidium minutissimum. Cocconeis spp. was also
common in the guts of these animals. Achnanthidium minutissimum is a
broad spectrum species often recorded as abundant in epilithic diatom river
studies in Ireland, Britain, Europe, Canada and the US (Ní Chatháin, 2002;
Sabater et al., 1988; Tomas and Sabater, 1985; Cox, 1990; Dixit and Smol,
1994; Pan et al., 1996; Rott et al., 1998). It is a ubiquitous species, one that
generally dominates diatom assemblages on cobbles in rivers of varying
water quality, therefore it is not a great indicator of any particular
environmental condition.
It does however seem to be more abundant in high status rivers, which was
also found in studies carried out during the RIVTYPE project in Ireland in
2002 and 2003 (Ní Chatháin et al., 2004). Empty filaments of green algae
were also found in the gut contents of Ecdyonurus venosus in the Castlebar
River but could not be identified with confidence due to the effects of the
digestion process on its structure. The impacted sites also contained an
abundance of diatoms. Achnanthidium minutissimum and Achnanthidium
lanceolata were also found in the guts of larvae sampled from the impacted
sites but were not as abundant. The most abundant diatom present in the gut
of one animal sampled from the Robe River (impacted site) was Reimeria
Sinuata. Overall the diversity of diatoms found in the high status and
impacted sites were quite similar. Based on the small number of animals
112
dissected in this spring study, it is very difficult to say whether there was a
difference between the high status and the impacted samples.
3.4.2 Analysis of stone scrapings
The following groups were quantified from the stone scrapings using a
combination of both the Lugols and DAPI stains: diatoms, filamentous
green algae, plant detritus and inorganic debris. This technique did not stain
biofilm matrix, coccoid green algae, algal agglomerates and coccoid
bacteria as effectively as using DAPI stain alone and unfortunately these
items could not therefore be included in the percentage compositional
breakdown of the stone scrapings for comparison purposes. A summary of
the percentage composition of the stone scrapings taken from a pooled
sample on either side of the experimental divide on 21st July and 9
th
September 2003 is outlined in Table 3.9.
Table 3.9 The mean percentage composition of particulate material in the
stone scrapings observed on the 21st July and 9
th September 2003.
21st July
2003
21st July
2003
9th September
2003
9th September
2003
Food category Treated
section
Control
section
Treated section Control section
Diatoms 17% 24% 32% 32%
Filamentous green algae 2% 10% 2% 6%
Inorganic debris 23% 10% - 3%
Plant detritus 58% 48% 66% 59%
Unidentified particles - 8% - -
3.4.2.1 Stone scraping results ‘pre-nutrient addition’ (21st July 2003)
The mean percentage composition of the particulate material in the stone
scrapings taken in the treated and control sections of the split-stream
experiment on 21st July 2003 (pre-nutrient addition) are graphically
represented in Fig. 3.15, 3.16.
113
Fig. 3.15 Mean percentage composition of particulate matter in the stone
scrapings in the treated section of the Castlebar River on 21st July 2003.
Fig. 3.16 Mean percentage composition of particulate matter in the stone
scrapings in the control section of the Castlebar River on 21st July 2003.
24%
10%
10%8%
48%
Diatoms
Filamentous green algae
Inorganic debris
Unidentified particles
Plant detritus
17%
2%
23%58%
Diatoms
Filamentous green algae
Inorganic debris
Plant detritus
114
3.4.2.2 Stone scraping results ‘post-nutrient addition’ (9th
September
2003)
The mean percentage composition of the particulate material in the stone
scrapings taken from both sides of the experiment on 9th
September 2003
(post-nutrient addition) are presented in Fig. 3.17 and 3.18.
Fig. 3.17 Mean percentage composition of particulate matter in the stone
scrapings in the treated section of the Castlebar River on 9th
September
2003.
Fig. 3.18 Mean percentage composition of particulate matter in the stone
scrapings in the treated section of the Castlebar River on 9th
September
2003.
The mean percentage composition of particulate material in the treated and
control sections of the experiment on the 21st July 2003 (post-nutrient
addition) appeared to be similar to those observed on the 9th
September
2003 (pre-nutrient addition). The particulate material in the stone scrapings
taken from the treated and control sections of the split-stream experiment on
the 21st July 2003 contained 17% and 24% of diatoms respectively (Table
3.9). The most common diatom found in these samples was Navicula spp.
32%
2%66%
Diatoms
Filamentous green algae
Plant detritus
32%
6%
3%
59%
Diatoms
Filamentous green algae
Inorganic debris
Plant detritus
115
Filamentous green algae represented 2% and 10% of the material studied
respectively. Plant detritus represented a large proportion of the total
particulate area examined in these samples, accounting for 58% and 48% of
the samples in the treated and control sections respectively. The particulate
matter found in the stone scrapings from the treated and control sections
contained 23% and 10% inorganic debris respectively. Unidentified
particles make up 8% of the sample examined in the control section, while
there was none found in the treated side.
The mean percentage composition of diatoms was higher in both sections of
the study in September 2003, where they represented 32% of the total
particulate material in each sample analysed (Table 3.9). Plant detritus was
present in large amounts, accounting for 66% and 59% of the samples
studied in the treated and control, respectively. Inorganic debris was found
in the control section of the study corresponding to 3% of the material in the
sample.
The material found in the stone scrapings did not fluoresce well, as
explained earlier, due to the preservative Lugols iodine interfering with the
DAPI stain which is also evident in the photomicrographs illustrated in Fig.
3.19A-F. The algal cells, in particular filamentous green algae, coccoid
green algae and algal agglomerates, do not fluoresce as effectively when
stained initially with Lugols iodine (Fig. 3.19D, E and F). A claw from an
early invertebrate instar was found in the stone scrapings taken from the
treated section of the split-stream experiment on 9th
September 2003 (Fig.
3.19A).
116
Fig. 3.19A 100x Invertebrate claw found in
the stone scrapings taken from the treated
section of the split-stream experiment on 9th
September 2003.
Fig. 3.19B 400x Plant detritus and
unidentified particles in a sample of stone
scrapings taken from the control section of
the experiment on 9th
September 2003
Fig. 3.19C 200x Filament of algae stained
using Lugols iodine highlighting the brown
chlorophyll taken from the stone scrapings in
the control section of the split-stream
experiment on 21st July 2003.
Fig. 3.19D 200x Filament of algae taken
from a stone scraping from the treated
section of the split-stream experiment on 21st
July 2003. The sample was preserved in
Lugols and subsequently stained with DAPI.
Fig. 3.19E 200x Filament of algae and
unidentified mass to the left. Material did not
fluoresce as effectively as when using DAPI
stain alone and appeared dull due to the
Lugols in the sample.
Fig. 3.19F 400x Plant detritus and algal
agglomerate. Nuclei and chloroplasts are not
fluorescing due to the Lugols staining the
cells.
117
3.4.2.3 Algal composition in pre- and post-nutrient addition samples
The algal material in the stone scrapings was examined quantitatively to
compile a list of taxa present in each of the four samples investigated. The
mean percentage composition of the algal taxa found in the stone scrapings
(15 fields viewed per sample) are summarised in Table 3.10. The mean
percentage composition of each algal species found in each sample studied,
are graphically represented in Figs. 3.20 to 3.23. Examples of the algal taxa
found in the stone scrapings are illustrated in Figs. 3.24A-F.
118
Table 3.10 Mean percentage composition of algal taxa in four samples
obtained from stone scrapings in the Castlebar River on two separate
occasions during 2003.
21st July
2003
21st July
2003
9th
September
2003
9th
September
2003
Algal list Control
section
Treated
section
Control
section
Treated
section Fragilaria spp. 24% 55% 35% 22%
Navicula spp. 12% 18% 3% 9%
Cocconeis spp. 30% 12% 14% 6%
Meridion spp. 11% - 13% 7%
Cryptomonas spp. 14% - 17% 39%
Cyclotella spp. 5% - 4% 2%
Gomphonema spp. 3% - 3% 4%
Trachelomonas spp. 1% - 4% 2%
Melosira spp. - 3% 1% 1%
Anabaena spp. - 8% - -
Oscillatoria spp. - 4% - -
Cymbella spp. - - 6% 4%
Oedogonium spp. - - - 3%
Mougeotia - - - 1%
Fig. 3.20 Mean percentage composition of epilithic algal taxa in the control
section of the Castlebar River on 21st July 2003.
Fig. 3.21 Mean percentage composition of epilithic algal taxa in the treated
section of the Castlebar River on 21st July 2003.
24%
12%
30%
11%
14%
5%3%1%
Fragilaria spp
Navicula spp
Cocconeis spp
Meridion spp
Cryptomonas spp
Cyclotella spp
Gomphonema spp
Trachelomonas spp
55%
18%
12%
3%
8%4%
Fragilaria spp
Navicula spp
Cocconeis spp
Melosira spp
Anabaena spp
Oscillatoria spp
119
Fig. 3.22 Mean percentage composition of epilithic algal taxa in the control
section of the Castlebar River on 9th
September 2003.
Fig. 3.23 Mean percentage composition of epilithic algal taxa in the treated
section of the Castlebar River on 9th
September 2003.
35%
3%
14%1%13%
17%
4%
6%
3%4%
Fragilaria spp
Navicula spp
Cocconeis spp
Melosira spp
Meridion spp (fragment)
Cryptomonas spp
Cyclotella spp
Cymbella spp
Gomphonema spp
Trachelomonas spp
22%
9%
6%
1%
7%
39%
2%
4%
4%2%3% 1%
Fragilaria spp
Navicula spp
Cocconeis spp
Melosira spp
Meridion spp
Cryptomonas spp
Cyclotella spp
Cymbella spp
Gomphonema spp
Trachelomonas spp
Oedogonium spp
Mougeotia spp
120
Fig. 3.24A 200x Meridion spp.
(Bacillariophyceae) from the stone scrapings
taken form the treated section of the Castlebar
River on 9th
September 2003 (post nutrient
addition).
Fig. 3.24B 200x Oscillatoria spp.
(Cyanobacteria) taken from stone scrapings
in the control section of the Castlebar River
on 21st July 2003 (pre-nutrient addition)
Fig. 3.24C 200x Fragilaria spp.
(Bacillariophyceae) in the stone scrapings
from the control sections of the Castlebar
River on 21st July 2003.
Fig. 3.24D 200x Anabena spp.
(Cyanobacteria) found in the stone scrapings
taken from the treated section of the
Castlebar River on 21st July 2003.
Fig. 3.24E 200x Melosira spp.
(Bacillariophyceae) found in the stone
scrapings in the treated section of the
Castlebar River on 21st July 2003.
Fig. 3.24F 200x Navicula spp.
(Bacillariophyceae) common to all samples
taken on both sampling occasions during the
study.
121
The algal taxa found in the stone scrapings are all common species present
in Irish rivers and are listed in Table 3.10. The most common genera found
in this river were Fragilaria spp., Navicula spp., and Cocconeis spp., which
were present in all four samples on both sampling occasions. Fragilaria
spp. was the most abundant algal species in the river during the study. The
mean percentage composition of Fragilaria spp. in the treated section
dropped from 55% when sampled on 21st July (pre-nutrient addition) to 22%
when sampled on 9th
September (post-nutrient addition). The mean
percentage composition of this species in the control section on 21st July
was 24% compared to 35% in the same section on the 9th
September. The
abundance of Fragilaria spp. appeared to fluctuate between sampling
occasions. The mean percentage composition of Navicula spp. and
Cocconeis spp. were lower in the autumn samples (9th
September) in both
the control and the treated sections of the experiment, compared to the
summer samples (Table 3.10). On both sampling occasions, the mean
percentage composition of algal taxa in the control sections of the
experiment was quite similar (Fig. 3.20 and Fig. 3.22).
Some algal species were found only in the treated section of the
experimental divide post-nutrient addition and were not present in the same
section when sampled prior to the addition of nutrient (Table 3.10, Fig. 3.21
and Fig. 3.23). It is not evident why this should happen, but may be due to
the natural variation in the algal community located within the river system
or to differences between sampling locations. The blue green algae,
Anabaena spp. and Oscillatoria spp. were found in the samples taken from
the treated section on 21st July 2003. They were represented in small
amounts, accounting for 8% and 4% of the total sample taken on this
sampling date. Both of these genera are best known as ‘nuisance
organisms’ but as they were not found in any of the other samples
investigated during this study, in particular in the samples taken post-
nutrient addition, they do not indicate eutrophic conditions in the river.
They may, however, be favoured by the fact that the system appears to be
nitrogen limited as some species at least are nitrogen fixers. Cymbella spp.
was only present in the samples analysed post nutrient addition accounting
for 6% and 4% of the total algal taxa found in the control and the treated
122
sections respectively. This species was not present in the samples analysed
pre-nutrient addition. It is a large genus that contains many common genera
and some species occur in all water types so its presence/absence is not
indicative of the trophic status of this river. The algal taxon, Oedogonium
spp. was found on one occasion only. It was present in the sample taken in
the autumn (post-nutrient addition) in the treated section, representing 3% of
the total algal taxa in this sample. These algae are abundant in stagnant or
slow-flowing shallow waters and are most frequently found in midsummer
(Pentecost, 1984). Samples were only taken on two occasions, on both sides
of the experimental divide. Ideally, sampling frequency would need to be
increased to a weekly sampling regime.
123
3.4.2.4 Summary of results from epilithic scrapings
The stone scrapings were examined quantitatively using inverted
microscopy to compile a list of algal taxa. The scrapings were also
examined using the same epifluorescent technique that was applied to the
gut contents. Filamentous green algae, algal agglomerates and coccoid
green algae were identified in the stone scrapings using this technique but it
was impossible to categorise these into more precise taxonomic groups.
This epifluorescent technique did not detect the array of algal taxa that were
found in the stone scrapings, compared with that found when using inverted
microscopy.
The stone scrapings were firstly assessed by use of the mean percentage
composition of particulate matter, which was estimated using the same
technique as was applied to the gut analysis specimens. As mentioned
previously, the preservative Lugol’s stains algal cells a dark brown but it
can mask chlorophyll fluorescence (Gifford and Caron, 2000). As a result,
not all of the particulate food categories found in the gut contents of the
Ecdyonurus venosus specimens were identifiable in the stone scraping
samples. Plant detritus was again the dominant material found in all of
these samples. Proportions were slightly higher in September compared to
July, where the largest amount (66%) was found in the ‘treated’ section.
Diatoms were present in all samples and were particularly abundant in
September 2003, accounting for 32% of the total particulate matter in each
of the samples taken from the ‘treated’ and ‘control’ sections of the
experiment. Filamentous green algae were again present in reasonable
amounts in the stone scraping samples throughout the study. Overall, there
were no marked differences proportionally in the percentage composition of
particulate matter between the stone scraping samples.
Secondly, the mean percentage composition of epilithic algal taxa was
examined during the study. The dominant species of algae on the stones in
the river throughout the study were Fragilaria spp., Navicula spp. and
Cocconeis spp. with the most common being Fragilaria spp. The
percentage mean composition of all three species in the ‘treated’ section
124
sampled on 21st July was nearly double the amount found in the same
section on 9th
September. The stone scrapings taken in the autumn had half
the amount of Navicula spp. and Cocconeis spp. present when compared
with the summer samples, in terms of their percentage mean composition.
The dominant diatom found in the guts of Ecdyonurus venosus in this river
was Navicula spp. albeit diatoms as a group were a relatively small
proportion of the total food items found in the gut of Ecdyonurus (2% and
7% in July and September respectively). These findings suggest that
grazing pressures by this mayfly may have reduced Navicula spp. as its
abundance increased on the substrata in the autumn months. It is also very
important to note that the relative amounts of diatoms in the gut contents
studied vs the stone surfaces seems to be lower indicating an avoidance of
diatoms by Ecdyonurus. They also suggest that grazing on Fragilaria spp.
and Cocconeis spp. increased, owing to the reduction of their proportions in
the autumn as the diatom growth reached a peak. Due to the absence of
these diatom species in the gut contents of Ecdyonurus venosus, it suggests
that they were grazed on by other macroinvertebrates in the river system.
Alternatively, it may be the result of changes in taxonomic composition
occurring in the river caused by direct manipulation of the river using
phosphorus but due to the small sample number investigated in this study,
this must remain as speculation.
Interestingly, Cocconeis spp. is often considered to be a grazer resistant
diatom species (Pan and Lowe, 1994). This adnate diatom is commonly
present and often dominant in algal communities grazed by caddisflies
(Lamberti et al, 1987; Steinman et al, 1987), mayflies (Colletti et al, 1987;
Hill and Knight, 1987), or snails (Steinman et al., 1987; Lowe and Hunter,
1988). Cocconeis spp. and Navicula spp. may have experienced nutrient
limitation and light limitation causing a reduction in their abundance, due to
the growth of other overstory species like Fragilaria spp. Cocconeis spp.
can secrete a tough mucilage which can secure the firm attachment of the
entire valve to substrate (Patrick, 1948) making grazing difficult. This may
account for the lack of Cocconeis spp. in the gut contents of Ecdyonurus
venosus and its abundance in the stone scraping samples. Fragilaria spp.
125
may also be difficult for Ecdyonurus venosus to graze on due to the
physiognomy of its structure.
3.4.3 Comparison of Ecdyonurus Gut Contents with Stone Scrapings
Fig. 3.25 Comparison of Ecdyonurus gut contents and stone surfaces for
those items that could be identified in both sets of samples: diatoms,
filamentous green algae, inorganic detritus, plant detritus and unidentified
particles.
On a sample basis, the mean percentage composition of diatoms in the stone
scrapings was higher than that found in the gut contents of Ecdyonurus
venosus (Fig. 3.25). It is thought that direct sampling of the river substrate,
may have contributed to the higher proportion of diatoms being detected in
these samples on both sampling occasions. As the samples were taken fresh
from the cobble substrates, more diatoms and algae were sampled without
damaging the algal community structures. In addition to this, the content of
the material available on the stone surfaces, during a particular feeding
episode, would have been a factor determining the overall composition of
food items in the invertebrate gut. Diatom growth is at its peak during the
spring an autumn months, owing to their abundance in the stone scrapings in
September compared to July 2003. Some of the stomachs may have been
altered due to the length of time between sampling and dissection.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Diatoms Filamentous
green algae
Inorganic
debris
Unidentified
particles
Plant detritus
Per
cen
tag
e re
pre
sen
atio
nGut Contents
Stone Surfaces
126
Fig. 3.25 attempts to make a comparison between the Ecdyonurus gut
contents and the epilithic material taken from the stone surfaces on which
they live and feed. The percentage breakdown is based on the average of all
the samples taken. The graph suggests that Ecdyonurus do not select
diatoms in that they are generally under represented in comparison with the
stone surfaces whereas inorganic debris seems to be present in greater
quantities than would be expected for a non-selective grazer. Plant detritus
dominates the samples of both gut contents and the stone surfaces with little
difference apparent.
A Chi-square analysis was carried out on a number of different treatment
combinations of the gut contents of Ecdyonurus venosus and the stone
scrapings taken from the Castlebar River during the summer of 2003 (Table
3.11) to establish if there were any similarities between the gut contents and
the food material scraped from the stone surfaces.
Due to the variability in the techniques used in analysing the gut contents
and the stone scrapings, only the following food groups were assessed in the
Chi-square analysis: Diatoms, filamentous algae, Inorganic detritus,
unidentified particles and plant detritus. As in Fig. 3.25, the analysis is
based on the average of all the samples taken from the gut contents and the
stone scrapings during the entire study and results are outlined in Table
3.11.
127
Table 3.11 Results of the Chi-square analysis carried out on different treatment combinations of the gut contents of Ecdyonurus venosus and the
stone scrapings taken from the Castlebar River during the summer of 2003.
Treated v
Control sections
Treatment combinations Chi-square
X2
Degrees of
freedom (df)
p-value
Treated section
Post-nutrient addition gut contents v Pre-nutrient addition gut contents
68.5
4
<0.001***
Control section Post-nutrient addition gut contents v Pre-nutrient addition gut contents
31.8 4 <0.001***
Treated section Post-nutrient addition stone surfaces v Pre-nutrient addition stone surfaces 37.3 4 <0.001***
Control section
1
2
Post-nutrient addition stone surfaces v Pre-nutrient addition stone surfaces 19.7 4 <0.001***
Treated section Pre-nutrient addition stone surfaces v Pre-nutrient addition gut contents 12.1 4 <0.05*
Post-nutrient addition stone surfaces v Post-nutrient addition gut contents
330.8 4 <0.001***
Control section Pre-nutrient addition stone surface v Pre-nutrient addition gut contents 15.9 4 <0.01**
Post-nutrient addition stone surfaces v Post-nutrient addition gut contents 84.7 4 <0.001***
3
4
Pre-nutrient addition gut contents treated section v Pre-nutrient addition gut
contents control section
5.1 4 0.3 ns
Post-nutrient addition gut contents treated section v Post nutrient addition gut
contents control section
96.1 4 <0.001***
Pre-nutrient addition stone surfaces treated section v Pre-nutrient addition stone
surfaces control section
35.4 4 <0.001***
5
6
7
8 Post-nutrient addition stone surfaces treated section v Post-nutrient addition stone
surfaces control section
6.5 4 0.164 ns
p-values: *P<0.05; ** P<0.01; *** P<0.001; ns – not significant
128
Significant differences in the gut contents were shown in the treated section
(p<0.001) and the control sections (p<0.001) both pre- and post-nutrient
addition (Treatment combination 1). There were also significant differences
found in the stone scrapings between the treated (p<0.001) and the control
sections (p<0.001) both pre- and post nutrient addition (Treatment
combination 2). While the p-values show significant differences in the gut
contents and the stone scrapings, these are merely reflecting the natural
variability in the diatoms and plant detritus in the river during the
experiment, as the differences were present both before and after the
nutrient was added to the experimental section. The significant differences
may also reflect the natural changes that occur in plant biomass in the river.
Significant differences between the gut contents of Ecdyonurus venosus and
the material found on the stone scrapings were found in the treated section
(Treatment combination 3) both pre-nutrient addition (<0.05) and post
nutrient addition (<0.001). Significant differences were also found in the
control section (Treatment combination 4) both pre-nutrient addition (<0.01)
and post nutrient addition (<0.001). There appears to be more diatoms and
plant detritus on the stone surfaces in both the control and treated sections
post nutrient addition. Diatoms are generally more abundant in the spring
and autumn months compared to the summer season which may explain the
increase in diatoms in the stone surfaces post-nutrient addition (September)
compared to the scrapings taken pre-nutrient addition which is contributing
to the overall increased significant value. Plant detritus was slightly higher
in the stone scrapings post-nutrient addition and may have been due to the
natural variation in plant material along the sampling gradient in the
experiment.
No significant differences in the gut contents were found between the
treated and the control sections pre-nutrient addition (Treatment
combination 5 – p=0.3). Significant differences in the gut contents were
found between the treated and the control sections post-nutrient addition
(Treatment combination 6 – p<0.001). There was more plant detritus found
in the guts of the larvae taken from the treated section (49%) compared to
the control section (35%) however the control section did contain slightly
129
more diatoms (9%) compared to the treated section (3%). Inorganic debris
was higher in the gut contents taken from the control section (41%)
compared to the treated section (21%). The significant difference found in
the gut contents between the treated and the control sections post-nutrient
addition appears to be attributed to the natural variation in the food items
available to Ecdyonurus during a particular feeding episode.
A significant difference in the content of the stone surfaces (p<0.001)
between the treated and the control section (treatment combination 7) was
evident pre-nutrient addition highlighting the natural variability in the
epilithic material on both sections of the experimental divide prior to the
addition of the nutrient. No significant difference (p=0.164) was found in
the content of the stone surfaces between the treated and the control sections
post-nutrient addition (Treatment combination 8).
130
3.5 Discussion
3.5.1 Introduction
Ecdyonurus is widely regarded as a sensitive taxon and indicator of good
water quality. A number of different hypotheses have been proposed for the
sensitivity of macroinvertebrates such as Ecdyonurus ranging from toxicity
effects, low oxygen concentrations, habitat siltation to food chain changes
due to enrichment.
In Ireland Ecdyonurus appears to be a particularly useful indicator of
eutrophication. In this context, therefore, the feeding regime of Ecdyonurus
is of interest. Thus, in this study, it was aimed to establish a method to
access the feeding habits of Ecdyonurus venosus by examining their gut
contents, in a small river in the West of Ireland. In a separate experiment, a
nutrient manipulation trial was set up in the Castlebar River, in an attempt to
artificially enrich one section of the river, during the summer of 2003
(Chapter 2). This provided an opportunity to assess whether such direct
manipulation had any obvious impact on the diet of Ecdyonurus. The
addition of phosphorus nutrient to one side of the split-stream experiment,
designed to raise the ambient nutrient concentration by 10 µgP/l, appeared
to cause an increase in the periphyton biomass, especially in the last three
weeks of the study. This experiment was used to investigate the diet of
Ecdyonurus venosus in this river before and after the nutrient manipulation.
It was hypothesised that the gut contents of the larvae would reflect the
changes in the algal biomass variation exhibited in the experimental site and
possibly reflect a variation in algal taxa between treated and control sections
of the experimental section.
Thus, the overall objectives were to establish what this key indicator species
was feeding on and to assess, the changes, if any, in its feeding regime
during the nutrient manipulation experiment.
131
3.5.2 Summary of findings
This study investigated the gut contents of Ecdyonurus over a number of
sampling occasions and initial findings give a good indication as to what
these larvae are eating. The study did not try to document the entire range
of feeding behaviour of the species, or to determine the nutritional
significance of the material ingested. It is also important to remember that
gut content studies only allow for determination of what is being consumed
and not what is being assimilated.
A preliminary examination of the gut contents of Ecdyonurus larvae
sampled during spring 2003, using the cold acid hydrogen peroxide
oxidation technique, was not very successful in isolating diatoms from the
specimens. Untreated guts were also examined and appeared to be more
effective in isolating the diatoms and the occasional filament of algae.
Other particles were found in the guts but were unidentifiable. This
technique however, was unable to detect any other items ingested that
would assist in increasing our knowledge of the diet of this mayfly.
The epifluorescent DAPI technique proved a very effective method in
isolating the dietary gut contents of these larvae. The results from the
preliminary investigation carried out on a few specimens, successfully
validated this method in isolating and identifying a variety of food
categories. The technique helped greatly in the categorisation of
Ecdyonurus into its functional feeding group in this study.
The stone scrapings were assessed and quantified using a combination of
both DAPI and Lugols stain. This method failed to stain biofilm matrix,
coccoid green algae, algal agglomerates and coccoid bacteria due to the
masking effect of Lugols on chlorophyll fluorescence. A new technique is
being developed to stain Lugols fixed plankton samples with DAPI. This
method is being prepared for publication (Struder-Kype et al., in prep) and
is outlined in the methods section in the Guide to UK Coastal Planktonic
Ciliates (2001). This allows the nuclear shape of microplankton to be used
as a diagnostic feature in routine plankton analysis.
132
On the basis of the findings from the gut analysis studies carried out in the
Castlebar River in 2003, the diet of the mayfly Ecdyonurus venosus mainly
consisted of epilithic algal tissue, plant particulate matter (detritus), biofilm
matrix and inorganic debris (mineral material). The Ecdyonurus guts
sampled in March 2003 included diatoms and sporadic findings of
filamentous green algae. It can therefore be classified as both a herbivorous
grazer and detritus feeder. Due to the presence of dead animal remains in
the guts on occasion, this genus may have an opportunistic feeding tendency
or it may be simply be that this type of material is ingested as the animal
grazes across the substrata. Interestingly, Moog (1995) also suggests that
this genus may function as both a grazer and a detritus feeder. The food
preference of these larvae appears to be strongly dependent on the food
particles present in the environment at a given time. Thus the diet of the
larvae of the studied species depends on the food resources of the
environment. From our findings, this mayfly feeds on those particles which
occur in the habitat in the greatest quantities or else those that are easily
accessible. This may explain the differences in the proportions of diatoms
in the guts of the animals investigated in March 2003 compared to those
studied from July-September 2003.
The majority of authors place aquatic insects in the trophic chain in the
position of primary consumers – herbivores, or secondary consumers –
predators (Cummins et al., 1966; Minshall, 1967; Hynes, 1970; Coffman et
al., 1971; Cummins, 1973; Shapas and Hilsenhoff, 1976; Gray and Ward,
1979). Ephemeropteran larvae are described by many authors as herbivores
(Minshall, 1967; Coffman et al., 1971; Edmunds et al., 1976) or as
herbivore-detritivores (Shapas and Hilsenhoff 1976; Gray, Ward, 1979).
According to Klonowska (1986), Ecdyonurus venosus larvae can be
considered mainly as grazers/scrapers feeding chiefly on diatoms and sessile
algae covering the mineral or organic substratum. The structure of its
mouthparts is well adapted for scraping and grazing the substrata. There
seems to be a relationship between the habitat of the mayfly and the type of
food, while the morphology and behaviour of the animal is also an
influencing factor.
133
Grazing studies carried out in a river in Switzerland by Wellnitz and Ward
(1998), suggest that Ecdyonurus to removes periphyton more evenly than
Baetis, producing a relatively thinner, cleaner looking mat. They imply that
this difference in architecture may have resulted from Ecdyonurus
“brushing” vs Baetis “scraping”. Their findings reveal that Ecdyonurus
harvest overstorey layers and that grazing by this invertebrate caused a
taxonomic shift to species that were richer in chlorophyll content. The
study hypothesised that the brushing mode of feeding employed by
Ecdyonurus would have greater influence on “tall” forms of algae whose
physiognomy would place then relatively high off the substratum which is
the ‘scraping and gathering’ zone; and indeed, stalked physiognomies were
influenced exclusively by Ecdyonurus.
In addition to this, diatom taxa may be predictably prostrate or erect with
respect to the substratum, however, the substratum to which they are
attached can be detritus or other algae within the periphytic matrix
(Stevenson, 1996). This action may also account for the apparently greater
proportion of inorganic debris in the gut of Ecdyonurus vis a vis the stone
scrapings as proposed by the work carried out in this study. The Swiss
study also reported that filamentous chlorophytes, for example, became
clearly more abundant in response to Ecdyonurus grazing. This increase
was due to Ulothrix and it may have resulted from Ecdyonurus cleaning
diatoms from this alga’s tough filament. Most overstorey periphyton
removed by Ecdyonurus appeared to be an amalgam of diatoms, silt and
detritus. These findings give an indication of the algal taxa preferred by
Ecdyonurus, which also agree well with the Castlebar findings. Unlike
Baetids, which make most of their large-scale movements by swimming
(Richards and Minshall, 1988), Heptageniids typically crawl with their
wide, flattened bodies pressed closely to the substratum. Non consumptive
losses of periphyton by mayfly grazing can be important (Scrimgeour et al.,
1991) and foraging Ecdyonurus undoubtedly detach substantially larger
amounts of periphyton than foraging Baetis. Lamberti et al. (1989)
suggested that removal of the periphytic overstorey can enhance the
productivity of underlying layers by easing nutrient and light limitation.
134
The dietary composition and feeding regime of four mayfly species were
examined in detail in a small stream in Poland (Klonowska, 1986). The
author found that it was those species that occurred most frequently in the
environment in a given season, which were ingested in the largest amounts.
An average of 32% of the food contents of Ecdyonurus venosus composed
of diatoms, the preferred types being Achnanthidium spp. (formerly known
as Achnantes sp.), Cymbella ventricosa, Meridion circulare, Navicula spp.
and Surirella ovata. These species were found in fairly large amounts in the
guts of almost all the specified larval age classes in all seasons of the year.
Other species were found only at particular times e.g. Achnanthidium
linearis var. pusilla (formerly know as Achnanthes linearis var. pusilla) in
winter and Navicula minima and Cocconeis placentula in winter and spring.
It was found by Klonowska (1986) that young Ecdyonurus venosus larvae
ate few diatom species, the number of diatom species ingested increasing
concurrently with their growth. He indicated that Ecdyonurus venosus
larvae preferred two diatom species in all seasons of the year i.e.
Achnanthidium minutissimum and Cocconeis placentula var. klinoraphis.
Other species of diatom were preferred only in specific seasons or by
younger or older larvae only, e.g. Achnantes lanceolata and Meridion
circulare. Findings from our studies on specimens collected in March 2003
showed that the dominant algal species found in Ecdyonurus venosus was
also Achnanthidium minutissimum. Cocconeis spp was again a common
taxon in the gut contents of specimens. Results from the investigations on
Ecdyonurus larvae sampled in the Castlebar River from July-September
2003, showed that the most common algal species found in their guts was
Navicula spp. It therefore appears that the food selected by Ecdyonurus
venosus for the diatom species ingested, are connected to a considerable
degree with the presence in the habitat in a given season. It may be that in
certain cases this selection is also brought about for purely mechanical
reasons – a greater ease in scraping some diatom species off the substratum
than others.
135
Due to the increase in algal biomass in the treated section of the nutrient
manipulation experiment towards the end of the study, it was hypothesised
that there may be a change in the algal taxa between the control and the
treated sections. This type of transformation was not evident, however, in
the stone scrapings (Table 3.11). As mentioned in the results section, the
only notable difference in the gut contents between summer and autumn
sampling, was the absence of diatoms in the guts examined in July, which
were present in the larval stomachs in September. This may be attributed to
the greater numbers of diatoms present naturally in rivers during the spring
and autumn seasons and was unlikely due to the addition of nutrient during
the manipulation experiment. Studies carried out by Paul and Duthie (1989)
on algal species responses to nutrients have shown that the growth of
overstorey algae became the most important component of community
change while understorey algal species density remained largely unchanged
as algal communities developed. They suggest that regulatory forces may
vary at the population level through the experiments. For example,
understorey species (adnate diatoms like Cocconeis and Achnantes) may
experience nutrient limitation and then light limitation due to growth of
overstory species (erect diatoms such as Synedra) (Meulemans, 1987; Paul
and Duthie, 1989).
A comparison was made between the Ecdyonurus gut contents and the
epilithic material taken from the stone surfaces on which they feed in the
Castlebar River in 2003. The diversity of algal taxa observed in the stone
scraping samples was not reflected in the gut contents of the Ecdyonurus
larvae. The results may be explained by the suggestion that Ecdyonurus
appears to feed on the upper layers of epilithic material rather than the forms
closer to the stone surfaces (Wellnitz and Ward, 1998).
In order to establish the feeding regime of Ecdyonurus in Ireland and to
investigate whether there was a temporal variation in the composition of the
diet, a more in-depth study like the structure of the following investigation,
could be undertaken. Ledger and Hildrew (2000) examined 15-20
individuals per month, over a 12-month period to establish the diet and
feeding routine of three nemourid stonefly species. Their findings suggest
136
that they ingested a large quantity of algae and biofilm in spring and winter
when substrata were cleared by flushing flows, but consumed fine
particulate organic matter and associated microflora in later summer and
autumn when discharge fell and these materials accumulated. They
summarised that, nemourids were strong opportunistic feeders that switched
their diet according to the availability of attached (biofilm) and loose
(FPOM) resources in Lone Oak over the study period. Temporal variation
in the food consumed by invertebrates was also observed by Chapman and
Demory (1963), working in two small Oregon streams. The possibility of
seasonal resource partitioning by macroinvertebrates is often forgotten by
aquatic ecologists who assume obligate trophic function for all taxa in a
given functional feeding group (Mihuc and Toetz, 1994).
The present study provides information on the feeding preferences of
Ecdyonurus in Ireland for the first time. They appear to be relatively
indiscriminate grazers of the upper storey layers of the periphyton and
associated detritus. There appeared to be a selection for non-diatom species
although diatoms were present in the diet especially in the Spring and
Autumn.
The hypothesis that Ecdyonurus would show diet changes in response to
mild phosphorus enrichment of a river was not supported by the results;
perhaps because no obvious changes in periphyton species communities
were not demonstrated either.
137
Chapter 4
Life history studies of the Heptageniidae family in five high
status rivers in the West of Ireland.
4.1 Introduction
To date the life history of Ecdyonurus has not been studied in detail in
Ireland and information on its larval growth of this genus is limited. A more
complete knowledge of the life cycles of this genus is required in order to
improve the sensitivity of the EPA Q-value System in the light of the
requirement for broader ecological assessment under the EU Water
Framework Directive. In particular, a more detailed knowledge of the life
cycles of the four most common Irish Ecdyonurus species would be
beneficial in helping to interpret the impact of pressures on the ecological
status of rivers. The peak pressures that occur during low-flow warm
summer periods are of particular interest in this respect.
Many other indicator species aestivate during the summer months – some
stoneflies and Rhithrogena semicolorata (Curtis), for example, and thus,
typically they are not found in benthic samples over much of the summer.
Generally speaking, however, at least one Ecdyonurus species is found
throughout the year. In cases where Ecdyonurus is not found during the
summer this is taken to be an indication of water pollution if habitat
requirements are met. Ecdyonurus may reappear later in the winter and
spring periods at such sites. It is important therefore to determine the
emergence periods and life cycle of the individual species of Ecdyonurus in
order to improve the interpretation of Q-values especially in cases where
Ecdyonurus larvae are not found during the summer months. By studying
the life cycles of Ecdyonurus species in a number of rivers with low levels
of environmental stress it is hoped to gain an insight into the likely impact
of pollution and other stresses on these indicator species at locations where
they have been eliminated from the faunal community.
138
Therefore, the aim of this study was to obtain quantitative information on
the life cycles of the Heptageniidae, in particular the genus Ecdyonurus, in
five high status rivers in the West of Ireland. Historically, the EPA
biologists and other freshwater biologists nationwide carry out routine
biological assessments for water quality purposes during the summer
months, a time when pollution pressures are at their greatest. It was
hypothesised at the commencement of the present study that the various
species of Ecdyonurus emerge in overlapping phases such that during the
summer months (when biological monitoring programmes are being
undertaken) larvae of at least one Ecdyonurus species will always be present
in the benthic riffle fauna of Irish rivers.
To understand and predict the response of organisms to variation and
change within and between freshwater ecosystems, one needs information
about their life histories (Power et al., 1988). Studies carried out by
freshwater biologists studying temperate stream sites over the course of a
year or two will have observed the seasonal succession of benthic species,
many appearing, others disappearing as the seasons progress. The sequence
of changes in size of individuals of various species over time will also have
been noted. These patterns reflect the growth and development of
individuals through their life cycle. A life cycle is the general sequence of
these morphological stages and physiological processes through which an
individual of a species passes during its life, effectively linking one
generation to the next (Giller and Malmqvist, 1998). The qualitative and
quantitative details of events associated with the life cycle make up the life
history, such as growth, development, dormancy, dispersal, number of
generations per year, etc. (Butler, 1984).
While the life cycle is essentially fixed for the species, the life history can
vary. For example, all aquatic Diptera pass through a life cycle involving
egg, larval, pupal and adult stages, but as far as life history is concerned, the
duration of stages, numbers of larval instars, activity of pupa and emergence
and flight duration of the adults can vary within and between populations
and species of flies. Two aspects of life history patterns are especially
important: voltinism which refers to the frequency with which the life cycle
139
is completed within a year and phenology which refers to the seasonal
timing of the various life cycle processes and the population synchrony of
these (Giller and Malmqvist, 1998). To determine life histories, biologists
try to follow the development and progression of individuals derived from
one reproductive period through the various life cycle stages or size classes
over time. These are based on linear measurements of body parts, usually
body length from tip of the labrum to the last abdominal segment or the
head capsule width etc. (Wallace and Anderson, 1996). The life history of
Ecdyonurus can vary between species and is often complicated by an
overlap of cohorts. A cohort is a group of individuals that were born at the
same time or in practice born over a short period. These cohorts must be
recognised and separated before estimates can be made for growth rates,
mortality rates and production (Elliott and Humpesch, 1980).
Within a species, prevailing conditions can determine how many
generations can be squeezed into the year. For example, the short summers
at high altitudes allow fewer generations than at lower latitudes. Univoltine
species, with a single generation each year, tend to be found only in certain
seasons. The life cycle of Ephemeroptera includes four stages: egg, larvae
or nymph, and two adult stages, subimago and imago. This cycle exhibits
incomplete metamorphosis, since it lacks the pupal stage (Engblom, 1996).
Extensive literature was summarised by Clifford (1982), included data on
718 life cycles for 297 species, the majority of which occur in Europe and
North America. He found that a large proportion of the Ephemeroptera
were univoltine. This type of life cycle exhibits a single generation each
year accounting for 60% of all ephemeropteran life cycles. A multivoltine
species, one that displays more than one generation per year, represented
30%, while 4% of all ephemeropteran life cycles were semivoltine (one
generation every two, or even three, years) and the remainder were variable.
The life cycles of most species vary slightly according to environmental
conditions (Elliott et al., 1988).
There is a great variation and flexibility in life history within and between
populations, including egg diapause, duration of larval stages and growth
rates, degree of synchrony in emergence and the number of generations per
140
year. This plasticity in life history can occur from intrinsic differences
between populations across the geographical range of a species (Giller and
Malmqvist, 1998). Temperature, for example, is inversely related to
duration of egg development in non-diapausing invertebrates and faster
development of arthropod instars occurs at higher temperatures. Nutritional
effects can also influence voltinism (Butler, 1984). In response to different
climates, therefore, many insect species may display more generations a
year at lower altitudes and latitudes. This is most apparent in widely
distributed species like Baetis rhodani, where due to changes in
environmental conditions, life cycles vary from being univoltine in northern
Europe to being bivoltine with both a winter generation and a summer
generation throughout most of Europe. It also presents itself as being
bivoltine with one winter and two summer generations in warmer streams of
Southern Europe (Clifford, 1982).
Genetic constraints may also limit voltinism, as in the mayfly Leptophlebia
which is always univoltine over a wide geographical and climatic range, but
variation does occur in most other ephemeropteran species (Brittain, 1982).
Life history patterns vary for the stonefly Nemoura trispinosa, from a
univoltine slow seasonal type to a univoltine fast seasonal type with
extended egg development dependent on maximum annual water
temperature (Williams et al., 1995). This species exhibits eurythermal egg
development – the ability to show major differences in the number of
degree-days needed for egg development among local populations of the
same species (Lillehammer et al., 1989). This allows for species to switch
between one- and two-year generation times depending on the local
temperature range and food supply.
The basic components for life history plasticity are prolonged hatching and
emergence periods and a wide range of larval stages at any one time. This
distributes the life cycle stages over time and thus decreases the risk of
eradication by short-term catastrophes (Dietrich and Anderson, 1995).
Drought during the summer is predictable and life cycle adaptations such as
drought-resistant eggs ensure rapid recolonisation following a very dry
spell. However, unpredictable winter or spring droughts do not allow for
141
such adaptations and hence the great value attached to life history plasticity
(Giller and Malmqvist, 1998). Coexisting caddis in the Glenfinnish River in
Ireland, demonstrate a wide variability in life history flexibility
(Sangpradub, Giller and O’Connor, 1999). These variations in rate of
development, including the ability to overwinter in different larval stages
and the asynchronous, extended flight periods, give increased flexibility to
cope with year-to-year differences in weather and unpredictable
disturbances, so called ‘spreading of risk’ (c.f. den Boer, 1968).
Life history parameters are also part of the framework underlying
‘ecological plasticity’ in mayflies (Brittain, 1991) and basically control the
ability to withstand both natural and man-made ecological disturbances.
The different life cycles of the Ephemeroptera (Landa, 1968; Sowa, 1975)
clearly influence both the completeness and comparability of samples.
Furthermore, among the Heptageniidae (which comprise about 45% of the
mayfly fauna of Central Europe) only the fully grown nymphs can be
reliably identified to species level (Bauernfeind and Moog, 2000).
Investigations that continue to neglect these considerations impair the
ecological and faunistic interpretations (Bauernfeind, 1998). As Reusch
(1985, 1995) and others have pointed out, monthly samples over at least one
year are the only means to obtain objective data on insect community
structure in streams and rivers.
From a scientific point of view, indicator organisms must be accurately
identified to species level (De Pauw and Vanhooren, 1983; Furse et al.,
1984; Hilsenhoff, 1987; Marten and Reusch, 1992; Moog et al., 1997; Resh
and Unzicker, 1975; Rosenberg et al., 1986). Categorisations of
macroinvertebrate life cycles are dependent on correct identifications.
Better taxonomy produces more accurate results and clearly increases the
precision of site classifications, improving the ability to detect subtle
changes in environmental quality (Lenat and Barbour, 1994). Modern
regional keys and revisions for identification of the Ephemeroptera are
available for most of Europe (Studemann et al., 1992; Bauernfeind, 1994;
Engblom, 1996). In Ireland, the keys used to identify the Ephemeroptera
are those published by the Freshwater Biological Association (Elliott et al.,
142
1988). Keys and descriptions published before 1980 contain valuable
information for the specialist but should be used with caution by the less
experienced: the evaluation of diagnostic characters has changed
considerably and many species have since been newly described
(Bauernfeind and Moog, 2000). Larval identifications require much care
because most key characters only apply to fully grown mature nymphs. No
reliable keys are available at present for early instars and misidentification
even at the genus or family level is possible.
Positive identification of species requires examination of all life stages for
most aquatic insects. However, in most cases, identification of species has
been based on nymphs or adults often with no association being made
between the two (Hynes, 1970; Smock, 1996, Merritt et al., 1996). Field
collecting of nymphs and adults in one location is an accepted method of
identifying all insect life stages but has the inherent problem with
discriminating between different species, especially if one has to rely on
immature nymphs for initial identification. An insect reared from an
immature stage to an adult, with the subsequent larval skin moult kept for
comparison, provides the definitive association. The rearing of mayflies can
be especially difficult because of the presence of a fragile subimago stage
which has characteristics different to those of the adult (Finlay, 2001).
The Heptageniidae possibly represent the most difficult family to identify
and instruction by trained taxonomists is recommended. In Ecdyonurus,
mature nymphs and male imagines can usually be correctly identified
(Bauernfeind, 1997; Hefti et al., 1989), but some training is essential. The
use of reference specimens is advisable to acquire the necessary experience.
Detailed information on the life cycle of Ecdyonurus in Ireland is quite
limited due to insufficient studies carried out in this area and to the
difficulty in the identification of this genus. This study addressed this
knowledge gap.
143
4.2 Study outline
This study was undertaken in order to characterise the life history of the
Heptageniidae family, with particular emphasis on the genus Ecdyonurus, in
a number of clean-water (high status) rivers in the West of Ireland.
144
4.3 Material and Methods
The site descriptions for each of the five high status rivers (Owengarve
River, Castlebar River, Brusna River, Dunneill River and Callow Loughs
Stream) are already described in detail in Chapter 1.
4.3.1 Experimental design and sampling regime
Quantitative macroinvertebrate sampling was carried out on a monthly basis
in all five rivers between July 2001 and October 2002 (see Table 5.4. in
Chapter 5 details of sampling dates). During each visit five individual
macroinvertebrate samples were taken per river using a 36cm x 36cm metal-
framed Surber sampler (500 microns mesh size). It was important that the
sampling area within each river included riffles, glides and margins so as to
fully represent all potential habitats within the system. A homogenous
sampling stretch was then chosen which included all relevant habitats and
cross-sectioned into 1m2
quadrats. Details of the length of the sampling area
for each of the ten river sites are outlined in the site descriptions in Chapter
1. Five quadrats were chosen randomly at each visit and a Surber sample
taken within each taking care to represent riffles, glides and marginal areas.
The substrate was carefully disturbed within the frame and all
macroinvertebrates were collected with particular care observed when
handling the Ecdyonurus specimens. All samples were preserved in 70%
IMS immediately on site for later identification.
Due to increased water levels in the rivers over the winter months, routine
Surber sampling had to be postponed until the levels receded. To continue
with the life history studies, Ecdyonurus specimens were collected by taking
a single kick sample in each river from November 2001 to January 2002
inclusive. Due to extreme weather conditions in February 2002, sampling
was halted completely and recommenced in March 2002.
The following chemical ‘quality’ parameters were also examined in each
river on a monthly basis: temperature and dissolved oxygen, % oxygen
saturation (both taken in the field), pH, conductivity, orthophosphate,
145
T.O.N., ammonia, chloride, BOD, alkalinity and colour. These are further
analysed in Chapter 5.
4.3.2 Laboratory studies: Identification and categorisation of monthly
samples
The Heptageniidae larvae were separated from all other macroinvertebrates
collected for the detailed life history studies. Larvae for the life history
studies were counted, identified and categorised into the three main genus
groups: Ecdyonurus spp., Rhithrogena spp. and Heptagenia spp. The body
lengths of individual larvae were carefully measured to the nearest 0.1mm
from the tip of the labrum to the end of the last abdominal segment, under a
Nikon Stereo Microscope. The larva of the genus Ecdyonurus and
Rhithrogena were then identified to species level using the Freshwater
Biological Association Key to Ephemeroptera (Elliott et al., 1988). The
genus Ecdyonurus was particularly difficult to identify so other keys were
also consulted, namely the ephemeropteran key compiled by Engblom
(1996). The Heptagenia specimens were identified to genus level only due
to time constraints.
One specific feature examined in the identification of the various species of
Ecdyonurus was the shape of the pronotum. The curvature differs between
species (Fig. 4.1 to Fig. 4.3). It tends to be more rounded and curved in
Ecdyonurus dispar (Curtis) and longer and less curved in Ecdyonurus
venosus (Fabricius). All genera of Ecdyonurus contain seven gills but
Ecdyonurus insignis (Eaton) is the only species that has a filament on all
seven gills, thus making it relatively easy to identify intact specimens (Fig.
4.4). The other three species are more difficult to separate and rely on an
examination of the curvature of the pronotum and setae on the prostheca.
These setae number less than 10 in the larva as Ecdyonurus torrentis
Kimmins whereas more than 10 setae are usually present in specimens
identified as Ecdyonurus venosus. This particular feature was very difficult
to examine and highly variable so it was therefore abandoned quite early on
in the studies.
146
Length frequency distribution charts were compiled for each month to
examine the life cycles of the individual species over the entire sampling
period (Appendices 4.1 to 4.20). Due to the difficulty in identifying the
small immature specimens of Ecdyonurus to species level, a separate
category designated “Ecdyonurus species” was also used. All specimens
that were not large enough to be accurately identified were placed in this
category, which are outlined in the life cycle histograms.
147
Fig. 4.1 Photomicrograph of the head and
thorax of Ecdyonurus venosus highlighting the
long thin curvature of the pronotum.
Fig. 4.2 Photomicrograph of the head and
thorax of Ecdyonurus dispar displaying a
shorter and more curved pronotum in
comparison to that in Fig. 4.1.
Fig. 4.3 Photomicrograph of the head and
thorax of Ecdyonurus insignis highlighting a
longer and wider pronotum than those found
in Ecdyonurus venosus and Ecdyonurus
dispar.
Fig. 4.4 Photomicrograph of the 7th
gill
found on the abdomen of Ecdyonurus
insignis containing a filamentous tuft.
Pronotum Pronotum
Pronotum
Filamentous tuft
148
4.3.3 Rearing experiments
In order to verify the identifications of the various species of Ecdyonurus, in
particular, those species that were more difficult to identify, such as
Ecdyonurus dispar, larvae were collected for rearing from a number of the
high status rivers during the summer of 2002. These rearing experiments
were established for taxonomic review rather than for other experimental
purposes. It was envisaged that this would be particularly helpful in
verifying species from rivers where several species could be present during
the same sampling period.
4.3.3.1 Rearing chamber
An uncovered glass tank measuring 55cm x 8cm x 40cm was used to house
the rearing chambers. The system ran on a continuous flow basis using an
electrically powered circulating pump and reservoir to circulate the water
supply back into the aquarium. The water level in the aquarium generally
remained at a level of approximately 5cm. As water was pumped around
the system, overflow was directed into the reservoir via a 5mm diameter
copper pipe that was fitted to one end of this aquarium. This ensured a
continuous supply of water to the aquarium on a 24-hour basis.
4.3.3.2 Data collection
Larvae for rearing were collected from the five high status rivers from June
to September 2002. From studying the literature, this phase represented the
main period for emergence of the genus Ecdyonurus. Late instar nymphs
with developing black wing pads were carefully removed from the substrate
with a paintbrush and placed in a bottle of stream water with a small stone
for attachment. The bottle was sealed and placed in a box, in this way live
larvae were successfully transported to the laboratory.
149
4.3.3.3 Laboratory rearing
The aquarium was filled with laboratory tap water until it reached the 5cm-
overflow pipe. Likewise, the reservoir was filled to the appropriate height
and the circulating pump attached. The water was allowed to circulate
overnight to oxygenate fully and to reach ambient room temperature and
also in order to reduce the dissolved chlorine burden of the original tap
water. The water temperature ranged from between 18°C to 21°C during
the hatching period.
Each chamber housed an individual specimen to ensure identification of the
individual and to enable verification by examining the larval skin and imago
left behind. The chambers (25cm x 10cm) were made of plastic mesh to
allow water to flow freely through the system. The live larvae were placed
in a petri dish of water and identified to species level if possible under a
stereomicroscope. Twigs and stones were collected from each site and
added to each chamber for the insect to use as a food source and as a
platform for emergence. They were then transferred quickly to a chamber,
the lid placed on top and secured tightly using bull clips. A nylon mesh was
tightly wrapped around the entire chamber using rubber bands to ensure no
animals escaped and to capture the adults as they emerged. Three large
aqua fizz aerators used to regulate airflow were also placed near the rearing
chambers in an effort to recreate stream flow conditions.
The aquarium was placed near a window exposing it to the normal
photoperiod of a 24-hour day but not direct sunlight. The larvae were
checked every day and the life stage of the individual noted. Once
emergence (or death) occurred the animals were removed and species and
sex determined by observation using a stereomicroscope. Empty chambers
were thoroughly washed and nymphs replaced as required. These new
nymphs were acclimatised to the controlled temperatures for a period of 20
to 30 minutes. After about 5-7 days, an algal slime had grown on the base
of the aquarium covering the stones and the rearing chambers. Larvae
appeared to be sensitive to this growth and quite a few died within 2-3 days
of placing them in the chambers. As a result, the aquarium was cleaned
150
every week with a phosphate free detergent (neutracon) to remove any algal
build up that may have affected the survival of these sensitive invertebrates.
151
4.4 Results
4.4.1 Physico-chemical data
The maximum, minimum, mean, median and standard deviation values for
the physico-chemical parameters are outlined and discussed in more details
in Chapter 5 (Appendix 5.1). The results indicate that the sites chosen were
of good water quality with low values for the indicators of organic pollution
and eutrophication, BOD, ammonia, total oxidised nitrogen (TON) and
phosphate (i.e. unfiltered MRP).
The temperatures observed in the rivers during the sampling period were
within the normal seasonal ranges. The mean pH values ranged from 7.7 to
8.0. Conductivity and alkalinity levels varied between the rivers reflecting
the different typologies, geology and fluctuation in the discharges among
sites. Chloride levels showed a narrow range in values in the high status
rivers. The mean BOD was low in all sites measuring approximately 0.8
mg/l O2 throughout the study.
There was, however, some suggestion of eutrophication impacts in the
Owengarve, the Dunneill and the Brusna Rivers. On occasion, the DO was
elevated (>120%) in these rivers (Appendix 5.1). Elevated levels of
ammonia were detected in the Owengarve River (0.04 and 0.07mg/l N) and
in the Dunneill River (0.11mg/l N) on occasion. High levels of unfiltered
MRP were also measured in some of the high status sites, particularly in the
Castlebar and the Brusna Rivers. TON concentrations fluctuated from 0.3
to 1.4 mg/l N in the Owengarve River and from 0.2 to 0.95 mg/l N in the
Brusna River.
The results outlined in Appendix 5.1 emphasise the changes and constant
fluxing experienced in the high status rivers during the life cycle studies.
4.4.2 Results from life history studies
Interpretations of the life history of each individual species studied are
outlined in Sections 4.4.2.1 to 4.4.2.5 below. The life cycles of the
152
Heptageniidae in the five high status rivers are graphically represented in
Figs. 4.5 to 4.28 and then some overall conclusions about the life cycles are
drawn for West of Ireland rivers as a whole.
4.4.2.1 Interpretation of the life history of Ecdyonurus venosus
Ecdyonurus venosus was present throughout the year in all five rivers
studied and appeared to show the emergence of two distinct broods. It is
therefore considered to be bivoltine.
Summer/Autumn 2001
The first brood emerged from August to September 2001 in the Owengarve
River (Fig. 4.5), from July to September 2001 in the Dunneill River (Fig.
4.6) and in August 2001 in the Brusna River (Fig. 4.7). Adults emerged
from July to August 2001 in both the Castlebar River (Fig. 4.8) and in
Callow Loughs Stream (Fig. 4.9). Hatching was evident in late autumn and
continued on into early winter in all rivers.
Spring 2002
The larvae of Ecdyonurus venosus grew slowly and steadily throughout the
winter months in all five rivers (overwintering larvae) and generally
emerged as the first brood from March to April/May 2002 (Fig. 4.6, 4.7, 4.8,
4.9) with the exception of the Owengarve River where they emerged until
slightly later from March to June 2002 (Fig. 4.5).
Summer 2002
The adults emerged throughout the summer months in the five rivers, laid
their eggs and matured very quickly to hatch in early June 2002. This fast
growing summer generation (second brood) grew rapidly to emerge as the
summer stock in July and August 2002 in four of the five rivers. Larvae in
the Castlebar River emerged on into September 2002 (Fig. 4.8). Ecdyonurus
venosus emerged in July 2002 in Callow Loughs Stream (Fig. 4.9).
The Brusna River represented the lowest abundance of Ecdyonurus venosus
of all the rivers studied and the life cycle here also appeared to be bivoltine.
Numbers were very sporadic throughout the sampling programme, which
153
made it quite difficult to interpret its life history with accuracy. A few
specimens were found in December 2001 having reached body lengths of up
to 15mm (Fig. 4.7). Larvae had not reached this size during December in
any of the other rivers studied, so it is difficult to say whether this
corresponds to an emergence period when such few numbers were found
during the sampling period. There is a possibility that these could be
univoltine specimens that emerged in the spring and didn’t manage to
emerge in August/September forcing them to overwinter.
Ecdyonurus venosus specimens ranged in size among river sites (Table 4.2
to 4.6). Throughout the emergence period of autumn 2001, the length of the
largest larvae in the Owengarve River ranged from 15-16mm. Those
measured in the other four rivers were 1-2mm shorter. A similar pattern
was observed the following autumn (2002) and all larvae were 1-2mm
smaller than those found the previous autumn (2001). The largest larvae
emerging in Callow Loughs Stream in spring 2002 measured from 14-
15mm, 1-2mm less than those found in the other four rivers. The
Owengarve River contained the largest specimens of Ecdyonurus venosus
among all sites studied while Callow Loughs Stream had the smallest larvae
throughout the study. There was no statistical difference (p=0.397) between
the mean size range of the largest larvae during the emergence period in all
rivers in 2001 compared to 2002.
In summary, Ecdyonurus venosus was present in all five rivers throughout
the year with the exception of brief episodes after emergence periods and its
life cycle appears to be bivoltine. Similar overwintering patterns with slow
growing larvae followed by rapid growth during the summer months were
also observed.
154
Fig 4.5 Life cycle of Ecdyonurus venosus in the Owengarve River from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Jun-02Nov-01 Dec-01 Jan-02 May-02 Jul-02Jul-01 Mar-02 Apr-02Aug-01 Sep-01 Oct-01
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Aug-02 Oct-02
= 1 individual
155
Fig 4.6 Life cycle of Ecdyonurus venosus in the Dunneill River from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Jul-01 Aug-01 Sep-01 Jan-02 Mar-02 Apr-02Oct-01 Nov-01 Dec-01 Oct-02Aug-02 Sep-02May-02 Jun-02 Jul-02
= 1 individual
156
Fig 4.7 Life cycle of Ecdyonurus venosus in the Brusna River from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Jul-02May-02Mar-02 Apr-02Jan-02Sep-01 Oct-01 Jun-02Jul-01 Aug-01 Nov-01 Dec-01
= 1 individual
157
Fig 4.8 Life cycle of Ecdyonurus venosus in the Castlebar River from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Aug-01 Oct-01Jul-01 Mar-02 Apr-02 May-02Nov-01 Dec-01 Jan-02 Jul-02 Aug-02 Sep-02 Oct-02
= 1 individual
158
Fig 4.9 Life cycle of Ecdyonurus species in Callow Loughs Stream from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Jul-01 Aug-01 Sep-01 Oct-01 Nov-01 Dec-01 Jan-02 Mar-02
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Apr-02 May-02 Jun-02 Sep-02 Oct-02Jul-02 Aug-02
= 1 individual
159
4.4.2.2 Interpretation of the life history of Ecdyonurus insignis
Ecdyonurus insignis was found in four out of the five rivers, being absent
from the Castlebar River entirely. The life cycle of this species showed one
distinct brood and for that reason was described as being univoltine with
overwintering larvae. It was present in the river benthos during the summer
months only.
Summer 2001
Larvae were not found in the Owengarve River during 2001. Larvae
emerged from the Dunneill River in July and August 2001 (Fig. 4.10) and in
August 2001 in Callow Loughs Stream (Fig. 4.11). One specimen was
found in Callow Loughs Stream in August 2001 which probably marked the
end of the emergence period for this species. Emergence in the Brusna
River occurred in August 2001 and September 2001 (Fig. 4.12) which
appeared to be slightly later than in the other rivers. As with Callow Loughs
Stream, only one specimen was found in the Brusna River in August 2001
signifying the end of the emergence period.
Summer 2002
Ecdyonurus insignis was found in the Owengarve River between June and
August 2002 (Fig. 4.13). Larvae emerged in the Dunneill River from May
to July (Fig. 4.10) and in the Brusna River from May to August 2002 (Fig.
4.12) while they emerged in Callow Loughs Stream from July to August
2002 (Fig. 4.11).
The length of the largest specimens of Ecdyonurus insignis in the four rivers
ranged from 11 to 15mm. The largest larvae were found in the Owengarve
River (14-15mm) while the smallest larvae were found in Callow Loughs
Stream (10-11mm). There was no significance difference (p=0.411) in
mean larval sizes between summer 2001 and summer 2002. In summary,
Ecdyonurus insignis was present in four out of the five rivers and displayed
a univoltine life cycle.
160
Fig 4.10 Life cycle of Ecdyonurus insignis in the Dunneill River from July 2001 to
October 2002.
Fig 4.11 Life cycle of Ecdyonurus insignis in Callow Loughs Stream from July 2001 to
October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Jul-02May-02 Jun-02Jul-01 Aug-01
= 2 individuals
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Aug-01 Aug-02
= 1 individual
161
Fig 4.12 Life cycle of Ecdyonurus insignis in the Brusna River from July 2001 to
October 2002.
Fig 4.13 Life cycle of Ecdyonurus insignis in the Owengarve River from July 2001 to
October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Jun-02 Aug-02Aug-01 Sep-01 May-02
= 1 individual
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Aug-02May-02 Jun-02 Jul-02
= 1 individual
162
4.4.2.3 Interpretation of the life history of Ecdyonurus dispar
Numbers of Ecdyonurus dispar were low and were found as mature larvae
only during the summer months. This species was found in all five rivers
during the sampling programme and had a univoltine life cycle with
overwintering larvae.
Summer 2001
Ecdyonurus dispar emerged in all five rivers in August 2001 (Figs. 4.14,
4.15, 4.16, 4.17 and 4.18). It is thought that recruitment of Ecdyonurus
dispar began in June/July 2001 when juveniles hatched and grew very
quickly to emerge a month or two later in August 2001.
Summer 2002
Larvae emerged during this season in the Owengarve River only in June
2002 (Fig. 4.14). The largest specimens found in the Owengarve River
were 11-12mm in August 2001 and since Ecdyonurus dispar was relatively
uncommon, however, it is possible that fully grown specimens larger than
12mm could have been missed due to their rarity in samples or due to rapid
growth in the period between two sampling dates at emergence time.
Recruitment more than likely commenced in late April/early May 2002 in
this river in order for the larvae to reach 14mm on emergence in June.
In summary, Ecdyonurus dispar was present in all five rivers studied and
had a univoltine life cycle. Numbers were low and it was the rarest of all
three species of Ecdyonurus found in these rivers during the study.
163
Fig 4.14 Life cycle of Ecdyonurus dispar in the Owengarve River from July 2001 to
October 2002.
Fig 4.15 Life cycle of Ecdyonurus dispar in the Dunneill River from July 2001 to
October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Aug-01 Jun-02
= 1 individual
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Aug-01
= 1 individual
164
Fig 4.16 Life cycle of Ecdyonurus dispar in the Castlebar River from July 2001 to
October 2002.
Fig 4.17 Life cycle of Ecdyonurus dispar in Callow Loughs Stream from July 2001 to
October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9 Jul-01 Aug-01
= 1 individual
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Aug-01
= 1 individual
165
Fig 4.18 Life cycle of Ecdyonurus dispar in the Brusna River from July 2001 to
October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Aug-01
= 2 individuals
166
4.4.2.4 Interpretation of the life history of Rhithrogena semicolorata
Rhithrogena semicolorata was present in all five rivers studied and showed
a univoltine life cycle with substantial numbers represented. The
overwintering larvae grew progressively during the winter months and final
instars emerged in March and April 2002 in the Owengarve River (Fig.
4.19). The emergence period was slightly longer in the Dunneill River (Fig.
4.20), the Castlebar River (Fig. 4.21), Callow Loughs Stream (Fig. 4.22)
and the Brusna River (Fig. 4.23) lasting from March to May 2002.
The larvae varied in size just before they emerged. The length of the largest
specimens in the rivers ranged from 8-13mm. The smallest of these were
found in the Owengarve River (8-9mm) while the largest were present in the
Dunneill River (12-13mm).
Eggs laid by the adults that emerged in March and April 2002 began to
hatch in September/October 2002 (depending on the river) creating a new
generation and numbers increased as expected during these months. This is
a pattern commonly observed by EPA biologists in the West of Ireland.
Rhithrogena is typically missing from river benthos particularly in July and
August with the earliest nymphs appearing in late August to early
September (McGarrigle pers. Comm.).
In summary, Rhithrogena semicolorata was found abundantly in all five
rivers and displayed a univoltine life cycle typical of this species.
167
Fig 4.19 Life cycle of Rhithrogena semicolorata in the Owengarve River from July 2001 to October 2002.
mm
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Jul-01 Sep-01 Oct-01 Dec-01
mm
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Apr-02 Oct-02Jun-02 Sep-02
= 20 individuals
Mar-02
168
Fig 4.20 Life cycle of Rhithrogena semicolorata in the Dunneill River from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Dec-01Oct-01 Nov-01Aug-01 Sep-01 May-02 Aug-02 Sep-02Jan-02 Mar-02 Apr-02
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9 May-02 Aug-02 Sep-02Jan-02 Mar-02 Apr-02 Oct-02
= 1 individual
169
Fig 4.21 Life cycle of Rhithrogena semicolorata in the Castlebar River from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Nov-01 Dec-01 Jan-02Jul-01 Sep-01 Oct-01
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Aug-02Jul-02Jun-02May-02Mar-02 Apr-02 Sep-02 Oct-02
= 1 individual
170
Fig 4.22 Life cycle of Rhithrogena semicolorata in Callow Loughs Stream from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Oct-01 Nov-01 Dec-01 Jan-02 Jun-02 Oct-02Mar-02 Apr-02 May-02
= 1 individual
171
Fig 4.23 Life cycle of Rhithrogena semicolorata in the Brusna River from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
0 - 0.9
Apr-02 May-02Oct-01 Nov-01 Mar-02 Oct-02Jun-02 Aug-02 Sep-02
= 1 individual
172
4.4.2.5 Interpretation of the life history of Heptagenia species
The Heptagenia specimens were not identified to species level, therefore,
the life cycle of Heptagenia was examined at the genus level only. Adult
specimens of Heptegenia sulphurea (Müller) were however identified from
the Owengarve rearing trials (see Table 4.7). The life cycle of the genus
was univoltine and was present in all five high status rivers.
The life cycle was very difficult to interpret in both the Dunneill and the
Brusna Rivers as larvae were poorly represented. They were found on only
one occasion in the Dunneill River (Fig. 4.24) and varied in size.
Heptagenia spp. was found in November 2001, January 2002 and March
2002 (Fig. 4.25) in the Brusna River only.
Adults emerged in the other three rivers from May to July 2002. The
Castlebar River contained the largest larvae and emerged as adults from
May to June 2002 (Fig. 4.26). Some sporadic specimens were still in the
benthos in July and August 2002. The larvae in Callow Loughs Stream
were poorly represented measuring 1-2mm less than those in the Castlebar
River and emerged from June/July 2002 (Fig. 4.27). A sporadic specimen
was found in August marking the end of the life cycle. Heptagenia emerged
in the Owengarve River from June to July 2002 (Fig. 4.28) with larvae of
similar size (11-12mm) to those found in the Castlebar River.
In summary, the life cycle of Hetagenia genus was univoltine. It was quite
difficult to interpret due to low numbers of larvae present in the rivers
during the sampling programme. The interpretation is therefore based on
findings from three (Owengarve River, Castlebar River and Callow Loughs
Stream) of the five high status rivers studied.
173
Fig 4.24 Life cycle of Heptagenia species in the Dunneill River from July
2001 to October 2002.
Fig 4.25 Life cycle of Heptagenia species in the Brusna River from July
2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Jun-02
= 1 individual
= 1 individual
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
0 - 0.9
Nov-01 Mar-02
174
Fig 4.26 Life cycle of Heptagenia species in the Castlebar River from July 2001 to October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9 Sep-01 Oct-01 Nov-01 May-02Dec-01 Jan-02 Mar-02 Apr-02 Oct-02Jun-02 Jul-02 Aug-02 Sep-02
= 1 individual
175
Fig 4.27 Life cycle of Heptagenia species in the Callow Loughs Stream from July
2001 to October 2002.
Fig 4.28 Life cycle of Heptagenia species in the Owengarve River from July 2001 to
October 2002.
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Oct-01 Nov-01 Apr-02 May-02Dec-01 Jan-02 Mar-02 Sep-02 Oct-02Jun-02 Aug-02
= 1 individual
Body Length (mm)
16.0-16.9
15.0-15.9
14.0-14.9
13.0-13.9
12.0-12.9
11.0-11.9
10.0-10.9
9.0 - 9.9
8.0 - 8.9
7.0 - 7.9
6.0 - 6.9
5.0 - 5.9
4.0 - 4.9
3.0 - 3.9
2.0 - 2.9
1.0 - 1.9
Jun-02 Jul-02
= 2 individuals
176
Table 4.2 Summary of life history of the Heptageniidae in the Owengarve River from July 2001 to October 2002.
Species
Emergence period
Total
Emergence period
Main
Length of
largest
larvae
First appearance
of small larvae in
large numbers
Over-wintering
larvae
Ecdyonurus
venosus
August-Sept 2001
March-June 2002
July-August 2002
August-Sept 2001
March-June 2002
August 2002
15-16mm
15-16mm
13-14mm
July 2001
Sept-Oct 2001
Aug-Sept 2002
yes
Ecdyonurus
insignis
June-August 2002 July 2002 14-15mm June 2002
yes*
Ecdyonurus
dispar
August 2001
June 2002
August 2001
June 2002
11-12mm
13-14mm
July-August 2001
June 2002
yes*
Rhithrogena
semicolorata
March-April 2002 March-April 2002 8-9mm September 2001
yes
Heptagenia
species
June-July 2002 June 2002 11-12mm September 2002
yes
* Assumptions based on findings from Elliott and Humpesch (1983), Macan and Maudsley (1968) and Wise (1980).
177
Table 4.3 Summary of life history of the Heptageniidae in the Dunneill River from July 2001 to October 2002.
Species
Emergence
Period
Total
Emergence
Period
Main
Length of
largest
larvae
First
appearance of
small larvae in
large numbers
Over -
wintering
larvae
Ecdyonurus
venosus
July-September 2001
March-April 2002
July-August 2002
July-August 2001
March-April 2002
July-August 2002
12-13mm
15-16mm
11-12mm
July 2001
Sept-Oct 2001
Sept-Oct 2002
yes
Ecdyonurus
insignis
July-August 2001
May-July 2002
July-August 2001
May-July 2002
13-14mm
11-12mm
May 2002
yes*
Ecdyonurus
dispar
August 2001 August 2001 10-12mm July 2001
yes*
Rhithrogena
semicolorata
March-May 2002 March-May 2002 12-13mm August and
September 2001
yes
Heptagenia
species
June 2002 June 2002 10mm June 2002
yes
* Assumptions based on findings from Elliott and Humpesch (1983), Macan and Maudsley (1968) and Wise (1980)
178
Table 4.4 Summary of life history of the Heptageniidae in the Castlebar River from July 2001 to October 2002.
Species
Emergence period
Total
Emergence Period
Main
Length of
largest
larvae
First appearance
of small larvae in
large numbers
Over-
wintering
larvae
Ecdyonurus
venosus
July-August 2001
April-May 2002
July-September 2002
July-August 2001
April-May 2002
July-August 2002
12-13mm
15-16mm
12-13mm
August 2001
June 2002
Sept-Oct 2002
yes
Ecdyonurus
dispar
August 2001 August 2001 12-13mm July-August 2001
yes*
Rhithrogena
semicolorata
March-May 2002 March-April 2002 10-11mm September 2001
yes
Heptagenia
species
May-June 2002 June 2002 12-13mm September 2001
yes
* Assumptions based on findings from Elliott and Humpesch (1983), Macan and Maudsley (1968) and Wise (1980).
179
Table 4.5 Summary of life history of the Heptageniidae in Callow Loughs Stream from July 2001 to October 2002.
Species
Emergence period
Total
Emergence period
Main
Length of largest
larvae
First appearance of
small larvae in
large numbers
Over-wintering
larvae
Ecdyonurus
venosus
July-August 2001
March-May 2002
July 2002
July-August 2001
March-May 2002
July 2002
11-12mm
14-15mm
10-11mm
Sept-Oct 2001
August-Sept 2002
yes
Ecdyonurus
insignis
August 2001
July-August 2002
August 2001
July-August 2002
10-11mm
11-12mm
June-July 2001
June-July 2001
yes*
Ecdyonurus
dispar
August 2001 August 2001 11-12mm Only found in Aug
2001 in very small
number
yes*
Rhithrogena
semicolorata
March-May 2002 April 2002 9-10mm October 2001
yes
Heptagenia
species
June-July 2002 June/July 2002 10-11mm October 2001
yes
* Assumptions based on findings from Elliott and Humpesch (1983), Macan and Maudsley (1968) and Wise (1980).
180
Table 4.6 Summary of life history of the Heptageniidae in the Brusna River from July 2001 to October 2002.
Species
Emergence Period
Total
Emergence Period
Main
Length of
largest
larvae
First appearance of
small larvae in large
numbers
Over-
wintering
larvae
Ecdyonurus
venosus
August 2001
March-April 2002
June-August 2002
August 2001
April 2002
June-July 2002
11-12mm
15-16mm
11-12mm
August-September 2001
Sept-Oct 2002
yes
Ecdyonurus
insignis
August-Sept 2001
May-August 2002
August-Sept 2001
May-August 2002
11-12mm
12-13mm
May 2002
yes*
Ecdyonurus
dispar
August 2001 August 2001 12-13mm August 2001
yes*
Rhithrogena
semicolorata
March-May 2002 April 2002 11-12mm October 2001
yes
Heptagenia
species
Insufficient data Insufficient data 6-7mm November 2001
yes
* Assumptions based on findings from Elliott and Humpesch (1983), Macan and Maudsley (1968) and Wise (1980).
181
4.4.2.6 Temperature variation among seasons
Seasonal variation in the monthly water temperatures from July 2001 to
October 2002 is given in Fig. 4.29. Measurements of the cumulative
degree-days for 2001, 2002 and 2003 are presented in Fig. 4.30.
Fig 4.29 Monthly changes in temperature in the five river sites from July
2001 to October 2002.
Water temperatures increased substantially from March to April 2002 (Fig.
4.29) which appears to coincide with the emergence of Ecdyonurus venosus
and Rhithrogena semicolorata larvae in all five rivers studied (Tables 4.2 to
4.6 and life cycle graphs below). Larvae of Ecdyonurus venosus,
Ecdyonurus insignis, Ecdyonurus dispar and Heptagenia spp. emerged at
different points during the summer seasons from July to September 2001and
subsequently from May to September 2002.
0
5
10
15
20
25
Jul.
'01
Au
g.'0
1
Sep
.'01
Oct
.'01
No
v.'0
1
Dec
.'01
Jan
.'02
Feb
. '0
2
Mar
.'02
Ap
r.'0
2
May
'02
Jun
.'02
Jul.
'02
Au
g.'0
2
Sep
.'02
Oct
.'02
Months [Jul.'01 - Oct.'02]
Tem
per
atu
re (
0C
)Owengarve River
Callow Loughs Stream
Dunneill River
Brusna River
Castlebar River
182
Fig. 4.30 Cumulative degree-days (air temperature) recorded at Straide
weather station for 2001, 2002 and 2003. Straide represents the nearest
station to the rivers studied in this investigation.
The cumulative degree-days data for 2001, 2002 and 2003 (Fig. 4.30) show
that 2001 was a cooler year compared to 2002. This is also reflected in the
changes observed in the water temperatures over the course of the study
(Fig. 4.29). Unfortunately, the sampling programme did not commence
until July and August 2001 so data from March to June (and July in the case
of the Brusna River) were not available. It was therefore not possible to
investigate whether larvae emerged later during this period in 2001
compared to 2002. It is important to note, however, that temperature, in
conjunction with other factors such as food quality and quantity, is an
important aspect controlling the life cycle of the Heptageniidae (Benke,
1984; Ward, 1992).
0
500
1000
1500
2000
2500
3000
3500
4000
0
28
56
84
11
2
14
0
16
8
19
6
22
4
25
2
28
0
30
8
33
6
36
4
Julian day
Cu
mu
lati
ve
deg
ree-
da
ys
2001-straide
2002-straide
2003-straide
183
4.4.3 Results from the rearing experiments
In general, mortality of the nymphs was quite high during the rearing trials.
A list of the species that successfully emerged into the imago stage is
presented in Table 4.7.
Table 4.7 Species list of successfully reared adults from the rearing trials.
River Date
collected
Species identified Number of
specimens/Sex
Owengarve River
24/06/03
24/06/03
04/07/03
04/07/03
11/09/03
Heptagenia sulphurea
Ecdyonurus insignis
Ecdyonurus insignis
Ecdyonurus dispar
Ecdyonurus dispar
2 x sex unknown
6 x Females: 2 x Males
2 x Female: 5 x Males
2 x Female
3 x Female: 5 x Male
Dunneill River
24/06/03
24/06/03
Ecdyonurus insignis
Ecdyonurus venosus
3 x Female
1 x Male
Castlebar River
18/06/03
28/07/03
10/07/03
Ecdyonurus venosus
Ecdyonurus dispar
Heptagenia sulphurea
2 x Male
2 x Male
1 x sex unknown
Brusna River
04/07/03 Ecdyonurus insignis 1 x Male
The duration of the emergence period varied throughout the trials ranging
from 1-7 days and generally depended on the stage of development of the
larvae when collected.
The emergence success of larvae of Ecdyonurus venosus was quite low in
comparison with larvae of Ecdyonurus insignis and Ecdyonurus dispar.
Larvae obtained from the Owengarve River appeared to have the greatest
survival rate in the aquarium and emerged quite successfully to the imago
stage. This was true in particular for larvae of Ecdyonurus dispar and
Ecdyonurus insignis. Due to the low numbers of Ecdyonurus larvae in the
Brusna River, there was difficulty in collecting adequate amounts for the
rearing trials from this river. Mortality was high in specimens obtained
from the Castlebar and the Dunneill Rivers, especially with specimens of
Ecdyonurus venosus. These tended to die within a few days of being moved
184
into the aquarium. This may in itself be an added indication of the relatively
high sensitivity of Ecdyonurus venosus to environmental stress in general.
185
4.5 Discussion
4.5.1 Introduction
Mayflies play an important role in almost all undisturbed freshwater
communities and their larvae frequently form a considerable part of the
material sampled during biomonitoring procedures. The genus Ecdyonurus
in particular, is widely accepted as a bioindicator for water quality and as a
key indicator species that forms an integral part of the EPA biological Q-
Value system for many years. It is important therefore to have a good
knowledge of the life cycle in order to be able to interpret the absence of
Ecdyonurus in particular. It is necessary to be able to state definitively that
at least one species of Ecdyonurus can be expected to occur in the riffle
benthos of unpolluted rivers, at any time of the year.
The EPA carries out routine biological assessments on rivers in Ireland on a
yearly basis, with an emphasis on summer months when pollution stresses
are expected to be at their highest. It was hypothesised that all species of
Ecdyonurus would be found during the summer season.
The overall objective of this study, therefore, was to establish the life cycle
of various species within the Heptageniidae family with specific emphasis
on the life cycle of the genus Ecdyonurus in five high status rivers in the
West of Ireland.
The Heptageniidae species examined in the five rivers in this study are
widely distributed in Europe. Ecdyonurus venosus, Ecdyonurus insignis,
Ecdyonurus dispar, Rhithrogena semicolorata and Heptagenia spp. are all
southern/central European species (Illies, 1978). The catchment areas of the
five rivers studied here occupy a somewhat intermediate latitudinal position
in Europe but located in a mild Atlantic-influence region of Ireland.
186
4.5.2 Interpretation of the life history of the genus Ecdyonurus
Ecdyonurus venosus was the dominant species in all five rivers studied in
this investigation. It was found throughout the study period with the
exception of a few months in some sites where it was absent prior to the
onset of hatching and new recruitment processes. The abundance of this
species in the benthos gradually increased rapidly initially, while small
larvae were being added to the population and maximum numbers usually
occurred towards the end of a hatching period. The life cycle of Ecdyonurus
venosus was bivoltine in all five high status rivers, displaying slow-growing
over-wintering generations that emerged in spring 2002 and fast-growing
summer generations that emerged during late summer and autumn of 2001
and 2002. It would be important to distinguish bivoltine populations from
overlapping univoltine populations that are just out of synch. The data
allowed us to say that we are not dealing with univoltine populations that for
example emerge in March/April and grow right through from then until the
following March/April with another group that emerge in September
growing through until the following September also.
Ecdyonurus venosus has been described in the literature as having two types
of life cycles (Fahy, 1973). It is described as being univoltine with
overwintering larvae, adults being found from April to July, sometimes to
October (Elliott, 1967; Landa, 1968; Thaibault, 1971; Sowa, 1975; Wise,
1980). On the other hand, studies carried out by Connolly and McCarthy
(1993), showed that the life cycle of Ecdyonurus venosus was univoltine in
the Corrib catchment, while it is capable of completing two life cycles per
year in continental Europe. However, the maximum sizes of nymphs of this
species recorded in the former study were up to 7mm longer than those
recorded in other European studies (Whitney, 1939; Hynes, 1961; Elliott,
1967 and Thaibault, 1971). Whelan (1980) described variations in the life
cycle of Ephemera danica populations in Irish lakes and attributed them to
differences in lake temperature regimes and other environmental factors.
Studies carried out by Fahy (1973) on the growth patterns of ten species of
Ephemeroptera in a small stream system in the West of Ireland (the
187
Altahoney system) described the life cycle of Ecdyonurus venosus as being
bivoltine. He also compiled emergence data or more accurately, flight
period data, by setting Mundie trappings, sweep netting the bankside
vegetation and beating tray collections from bankside trees. This
contributed to a more accurate description of the life cycle of this species.
Wise (1980) described the life cycle of Ecdyonurus venosus in a
Northumbrian River in England as univoltine. Elliot (1967) also noted that
the life cycle of this species in a tributary of the East Dart, in central
Dartmoor, was univoltine. This was in contrast to the findings of Rawlinson
(1939), in a study on the River Alyn, who found that Ecdyonurus venosus
had a fast growing summer generation in addition to a slow-growing winter
generation (bivoltine).
The emergence periods of Ecdyonurus venosus showed only slight
variations between the river sites studied. It is widely known that the
temperature together with the quantity and quality of the food available are
the main factors that effect the life history of aquatic insects (Benke, 1984;
Ward, 1992). The duration of the larval life of a species may vary with the
season of the year and also with its geographical distribution. The life cycle
of Ecdyonurus venosus herein described, showed a seasonal variation in the
length of larval life and was more than likely directly dependent upon
environmental conditions, particularly on temperature and food supply. It is
reasonable to suggest that body sizes may therefore vary with local
conditions, which will determine the length of the whole aquatic period of
life and the season of emergence. This may account for the slight variation
in emergence patterns between the rivers studied in this investigation. As
temperatures increased in the rivers during this study in spring 2002 and
food became more abundant the over-wintered larvae began to grow more
rapidly. The availability of more food more than likely accounted for the
larger body sizes during this period.
The fast growing summer stock of Ecdyonurus venosus emerged slightly
earlier in 2001 in the Castlebar River, the Dunneill River, Brusna River and
Callow Loughs Stream, compared to the Owengarve River. The slow-
growing winter generation emerged during approximately the same period
188
in spring/summer 2002 but extended into June in the Owengarve and into
May in the Castlebar River and Callow Loughs Stream. Another fast
growing summer generation emerged in the five river sites in late
summer/early autumn of 2002 when the average larvae had reached only 11-
12mm. It is postulated that the principal driving factor determining whether
emergence can occur or not is body weight rather than body length and that
the slow-growing overwintering larvae lose weight during the cold winter
period even if they perhaps undergo an additional moult allowing them to
grow significantly longer than the summer generation. When they emerge in
March April, however, they have just reached the body weight necessary to
emerge and reproduce as adult insects.
The slow growing, over-wintering generation that emerged in spring 2002
seemed to grow to larger sizes compared to the fast growing summer stock
of both 2001 and 2002, although there were no statistical differences
observed. Even though egg development was not studied in this
investigation, it was assumed from the interpretation of the findings from
these studies that the majority of the summer eggs hatch within one or two
months and similarly the eggs laid by the late-autumn adults hatch fairly
rapidly and overwinter as larvae and taking from eight to ten months to
complete their growth. The winter stock appeared to have developed from
eggs oviposited in the autumn that either hatched in late autumn or remained
as eggs that slowly hatched over the winter into larvae with a slower rate of
growth caused by low temperatures. The rapid growth in spring and early
summer was no doubt due to the abundance of plant food and the higher
river temperatures, the slower growth, to cold and the scarcity of food. The
larvae probably browsed on stones covered with algae and plant debris
(Chapter 2) but during the cold months food is usually relatively scarce and
can be further depleted by the scouring action of the winter floods
(Rawlinson, 1939).
One of the major difficulties encountered in constructing the life cycles of
the genus Ecdyonurus was identifying accurately the small immature larvae
to species level. These unidentified juvenile larvae were assigned to the
“Ecdyonurus species” category in the length frequency histograms. It was
189
particularly difficult to identify small larvae of Ecdyonurus dispar as the
main diagnostic feature in differentiating between the species of
Ecdyonurus, namely the pronotum, was not fully developed in larval
specimens under 7-8mm in length. The larvae of Ecdyonurus insignis were
much easier to identify as all seven gills have a diagnostic tuft or filament
attached.
Even after the majority of the life cycles were constructed, it was still
difficult to assign these immature larvae to a particular species. This proved
even more complicated when the three different species of Ecdyonurus were
found in a sample at the same time of year, like that found in the Owengarve
River during the summer of 2002. Unfortunately, studies on the duration of
egg development of the genus Ecdyonurus was not carried out during these
investigations. There was an attempt to examine the flight periods of this
genus, particularly in the more productive systems but it proved
unsuccessful due the difficulty in finding adult specimens and time
constraints. It was abandoned after a number of weeks when no adults were
captured. Detailed studies on both the emerged adult populations and
knowledge of the development of eggs are essential to complete the full life
history of this sensitive indicator genus.
Ecdyonurus dispar emerged in all rivers during August 2001 only but
emerged again the following June 2002 in the Owengarve River owing to
the productive nature of this system. It emerged in August 2001 in all five
rivers and appeared to happen at a time when the majority of the other
Ecdyonurus species had already emerged or were about to reach the end of
their emergence period. It was hard to say whether larvae of Ecdyonurus
dispar grew over the winter months as the juveniles measuring under 8mm
in length were extremely difficult to identify. Larvae greater than 8mm in
length however were only identified in August 2001. The life cycle of
Ecdyonurus dispar in these high status rivers was univoltine.
Studies carried out by Macan and Maudsley (1968) in Lake Windermere in
Northern England and in the River Lune in 1979 and 1981 reveal a similar
life cycle. The larvae grew all through the winter and some had reached full
190
size by May or June, though specimens of all sizes were still present.
Observations on emergence patterns were recorded by Kimmins (1972) who
recorded a flight period from June to October. In June the number of small
nymphs increased and Macan and Maudsley (1968) interpret this to be the
appearance of a second generation. However this interpretation was found
to be incorrect because Humpesch (1980), on the rate of development of
eggs and larvae at different temperatures showed that these small larvae
could not have come from the eggs laid the same year; they could only have
come from eggs laid late during the previous season when the temperature
was falling to a level at which the development was slow. It is possible that
eggs laid by the earliest adults gave rise to a quick summer generation
emerging late in the season, but most of the populations achieve one
generation in a year. This was also observed during his studies in 1981.
The explanation favoured as to why the larvae were not present in the
samples taken in July are that the larvae were prevented from colonising the
superficial stones by the larger larvae of the other species like Ecdyonurus
venosus.
Studies carried out by Wise (1980), showed that Ecdyonurus dispar,
Ecdyonurus venosus and Ecdyonurus torrentis were all univoltine and that
Ecdyonurus torrentis and Ecdyonurus venosus were both winter-growing
species which emerged in the spring. Ecdyonurus dispar exploited the same
niche during the mid-summer period when the other species of the genus
were absent from the benthos. Macan (1957) found a similar cycle for
Ecdyonurus torrentis. However, he also found that growth and emergence
continued throughout the summer. Harker (1952) suggested that
Ecdyonurus torrentis had a cycle of three generations in 2 years but the
nymphs which hatched between August and April in the Coquet in Britain
are thought to belong to the same cohort. However, Macan (1970) stated
that Ecdyonurus dispar is absent from other lakes in the winter, though
often abundant for a brief period in the summer.
Macan and Maudsley (1968) also observed the absence of small (2mm) and
medium-sized larvae of Ecdyonurus dispar throughout the winter and in
spring in some English lakes, e.g. Ennerdale Water (Macan, 1970) and in
191
English and Central European rivers (Landa, 1968; Sowa, 1975; Macan,
1979; Wise, 1980) which led to their conclusion, that there was only a quick
summer growing generation. As far as the Ennerdale population was
concerned, the latter interpretation was found to be incorrect because there
was no evidence for an egg or larval diapause in winter and indeed, in 1977-
78 when this study was carried out, small larvae (2mm) of Ecdyonurus
dispar were collected during the winter. In May 1978, tiny larvae were also
captured and these could not have been progeny of adults emerging in 1978.
The absence of specimens in previous studies in this system may have been
due to an irregular distribution and a scarcity of larvae in some lakes.
Ecdyonurus insignis was found in four out of five of the high status rivers in
this study. It was completely absent from the Castlebar River. This species
displayed a univoltine life cycle. It emerged from the Dunneill River, the
Brusna River and Callow Loughs Stream towards the end of July and on
into August 2001. The larvae of this species usually emerged when the
numbers of larvae of other species of Ecdyonurus were low which usually
indicated the end of an emergence period. Emergence periods for
Ecdyonurus insignis varied slightly between the rivers. During 2001, this
species emerged mainly during August and September 2001. The following
year, in some of the rivers, the larvae began to emerge as early as in May
2002 while others emerged on into August 2002. With the exception of a
study carried out in the River Lune by Macan in the late 1970’s (1981), no
previous investigations on Ecdyonurus insignis have been discovered. In
his study, Macan found a generation of Ecdyonurus insignis during the
summer months only and described it as having a univoltine life cycle with
overwintering larvae. In central Europe this species overwinters in the egg
stage (Landa, 1968; Sowa, 1975, 1979). Adults of this species have been
found from May to October in the British Isles (Elliott and Humpesch,
1983).
As with Ecdyonurus dispar, Ecdyonurus insignis was found as mature
larvae only during the summer months, predominantly during the months of
May through to August. Again, there was no evidence of this species in the
rivers during the winter or spring seasons. The absence of larvae of both
192
Ecdyonurus dispar and Ecdyonurus insignis during the winter and spring
months suggests that these species develop quite differently to the more
common species Ecdyonurus venosus. No examination of egg development
of these species was carried out so it was difficult to ascertain whether they
overwintered as eggs and larvae. Interpretations are based on the
suggestions and material obtained from previous studies in the literature,
which were mostly based in Britain in particular those from Elliott and
Humpesch (1983) and Macan and Maudsley (1968). Findings from studies
carried out in Ireland by Wise (1980) also support the suggestion that
Ecdyonurus venosus and Ecdyonurus dispar overwinter as larvae. There
have been no further studies to date of this nature in Ireland prior to this
present investigation.
The absence of larvae of Ecdyonurus dispar during the winter and spring in
localities where other Ecdyonurus spp. occur as well and where a definite
pattern of the flight period of the latter has been observed, for example, in
the River Lune. This led Macan (1981) to the following hypothesis: in the
presence of larger larvae of another species, smaller ones are confined to the
deeper parts of the substratum and do not colonise the stones at the surface
until the larger have emerged. He suggests that while confined to the deeper
regions larvae grow less rapidly than when at the surface. This so called
‘size hierarchy effect’ (Brown, 1957; Ricker 1979) is a common feature in
trout hatcheries for fish having started at about the same size and it is
understood that when the food supply is limited the establishment of a social
hierarchy can result in an ‘adequate’ food supply and rapid growth for the
dominant fish, and a ‘less than adequate’ food supply and slow growth for
subordinate fish. When larger fish are removed, smaller ones recommence
rapid growth once again (e.g., Coates, 1980). In some situations, the growth
of smaller fish in a group of brown trout fry would not accelerate even under
the conditions in which an adequate supply of food was available to support
rapid growth by all (Peter, 1979).
Similar effects of intra-specific competition, shown for the freshwater
crayfish Oronectes virilis (Hagen) by Momot and Jones (1977), indicate
reverse relationship between growth rate and total standing stock biomass.
193
Variation in size of Ecdyonurus dispar of the same brood was observed in
the experiments of groups of larvae at 20, 14, 9°C, and similar observations
have been made by several other workers, e.g., Rawlinson (1939); Hunt
(1953); Trost and Berner (1963); Bohle (1978), for other ephemeropteran
species. It is not known what may have caused this variation and if the
latter occurs in the field as well. However, different growth rates of larvae
of the same brood maintained under the same environmental conditions may
explain the existence of a long summer emergence period in some species or
the split of the emergence period of an autumnal cohort into late autumn and
early spring one (Humpesch, 1981). The fact that species with a long flight
period tend to produce larger females and hence more fecund females early
in the season has been reported by several workers, e.g., Beneche (1972),
Elliott and Humpesch (1980). Sweeney and Vannote (1978), have proposed
that such changes in North American species are due chiefly to temperature
affecting adult size and fecundity by altering the larval growth rate and the
timing and rate of adult tissue development.
Information on hatching time is therefore essential for the recognition and
separation of cohorts of ephemeropteran species in the field, and cohorts
must be separated for accurate estimations of growth rates, mortality rates
and production (Humpesch, 1980). As a result of extensive field studies,
Landa (1968) determined the number of generations per year for several
ephemeropteran species by observing larval development. He assumed that
eggs of Ecdyonurus dispar and Ecdyonurus insignis passed through a
diapause stage, because newly-hatched larvae or small larvae could not be
found in the samples during the autumn and winter, long after the flight
period. Several other workers have made similar interpretations. Studies by
Humpesch (1980) indicates that the classifications of Landa (1968) are only
partially correct as there is no evidence of an obligatory diapause in eggs of
Ecdyonurus insignis (River Eden) and Ecdyonurus dispar (Windermere,
Lake Ennerdale). He surmised that the newly hatched larvae must be
present in autumn and winter in these rivers. There is some information on
the occurrence of young stages of Ephemeroptera and some workers have
pointed out that most of them live deep in gravel beds (e.g. Macan, 1958;
194
Tilzer, 1968) whereas most of the older larvae live nearer the surface.
Therefore, it may be difficult to find or catch the smaller larvae.
In all of the rivers investigated, the most common species present was
Ecdyonurus venosus and it was also the largest of the larvae. The suggestion
by Macan (1981) that in the presence of larger larvae of another species, in
this case Ecdyonurus venosus, smaller ones are confined to the deeper parts
of the substratum and do not colonise the stones at the surface until the
larger have emerged, appears plausible in explaining why both Ecdyonurus
insignis and Ecdyonurus dispar are only found in the benthos during the
summer months. It reinforces the idea that the larvae grew very slowly
during the winter months and began to grow steadily when the temperatures
rose during the spring and summer seasons to emerge in late summer. Their
absence from the benthos in winter months may also suggest that the eggs
overwintered and hatched into a fast growing summer generation.
Partitioning of emergence periods appeared to be evident in this
investigation, particularly in the Owengarve River as this river contained an
abundance of Ecdyonurus specimens. The winter generation of Ecdyonurus
venosus emerged between March and June 2002, Ecdyonurus dispar
predominantly emerged in June 2002 while the larvae of Ecdyonurus
insignis emerged in July 2002. Finally, the fast growing summer generation
of Ecdyonurus venosus emerged in August 2002. Each species appears to
divide itself into separate emergence periods during the summer months in
order to maximise survival rates. Depending on the number of Ecdyonurus
species present in the rivers, each individual species appeared to have
adapted their life cycles to emerge at the most suitable time, in order to
ensure the survival of the next generation. This phenomenon was also
evident in studies carried out by Macan (1981). He investigated the life
history of Ecdyonurus torrentis and Ecdyonurus dispar in the Lune River
and found them to be similar, except that, at the same temperature,
Ecdyonurus torrentis emerged earlier than Ecdyonurus dispar and its eggs
hatched earlier. He implies that this may account for the way in which the
two species divide the Lune between them in the manner observed.
195
Some other workers have found that the flight periods of the different
species in the same biotype may follow a definite pattern, e.g. adults of
Ecdyonurus torrentis appear first followed by Ecdyonurus insignis and
Ecdyonurus dispar (Landa, 1962). From these field studies, some authors
have suggested that these variations in the life cycle or this succession of
species may be due to different patterns of egg development or larval
growth e.g., it is suggested that the eggs of Ecdyonurus torrentis hatch about
one month after oviposition, while those of Ecdyonurus dispar and
Ecdyonurus insignis do not hatch until the following year (Landa, 1968;
Sowa, 1975). Again, this information underpins the suggestion from our
findings that different species in the same river system, like those observed
in some of the rivers investigated in this study, follow chronological
patterns of emergence periods in order to ensure the survival of each
individual species. Therefore, different patterns in life cycle and succession
of Ecdyonurus spp. can be partly explained by variation in their hatching
times at different temperatures and in different localities but further studies
need to be carried out in Ireland which would include egg development and
flight period studies.
196
4.5.3 Interpretation of the life history of Rhithrogena semicolorata
Rhithrogena semicolorata was found in the five high status rivers
investigated and was well represented at all sites. It clearly displayed a
univoltine life cycle where emergence periods ranged from three months
(March to May 2002), to two months (March to April 2002). The majority
of the final instars emerging during March and April 2002. It possessed a
slow-growing generation that over-wintered as larvae and grew steadily as
the temperatures increased in the spring. Larvae were absent from the
benthos after the emergence period for only one month in some rivers while
this was extended to up to three months in others (typically from June to
August) prior to the onset of another hatch. The eggs laid by the emerged
females began to hatch in September/October 2002 into young larvae
indicating that there was no embryonic diapause.
The larvae found in our studies varied in size prior to emergence. The
smallest larvae to emerge were found in the Owengarve River ranging from
8-9mm while the largest to emerge were captured in the Dunneill River
measuring between 12-13mm. Studies carried out by Fahy (1973) display
some variation from the typical pattern found by other authors, chiefly in
that there was a fall in the numbers of instars from March to April 1971.
Harker (1952) indicated a final instar size of 14mm and Macan (1957) final
instar size of less than 12mm. Hynes (1966) reported a final instar size of
less than 10mm. In the work described by Fahy (1973) the largest nymphs
were 13mm long at the beginning of the emergence period but later only
smaller ones could be found and the last to emerge were between 9 and
10mm long. He points out that the size difference may be the result of the
emergence of smaller individuals as the hatch progresses, a phenomenon
reported as widespread among the Ephemeroptera (Coleman and Hynes,
1970).
Eventhough the temperature in the rivers was only taken once a month (with
a gap of up to 4 weeks between measurements) there was a notable increase
of between 5-7°C in the water temperatures from March to April 2002. This
rise in water temperature appeared to signal the onset of the emergence of
197
Rhithrogena semicolorata in the rivers studied. Irrespective of what size the
larvae had already reached during this period, they began to emerge from
the rivers apparently triggered by the warmer waters (13-15°C) and ensuring
the survival of the next generation by emerging as temperatures were on the
increase.
Rhithrogena semicolorata occurs in South and Central Europe, Britain and
Denmark, but is not found in the North European region of the Soviet Union
or in Scandinavia (Sowa, 1975). With the exception of the study carried by
Fahy (1973) there have been no investigations into the life cycle of
Rhithrogena semicolorata in Ireland to date therefore comparative
information on the life history of this species is taken from studies carried
out in England and Switzerland. According to Elliott and Humpesch
(1980), nymphs of Rhithrogena semicolorata appear to grow very slowly
after hatching and the life cycle takes about one year from oviposition to
emergence of the adults. Studies by Wise (1980) in a Northumbrian river,
showed that the life cycle of Rhithrogena semicolorata was also univoltine,
possessing a slow-growing winter generation which commenced hatching in
early autumn and emerged during the spring. Nymphs were absent from the
benthos during mid-summer. The life cycle of Rhithrogena semicolorata in
the rivers studied in the present investigations reflected a similar pattern. In
studies carried out in a prealpine stream system in Switzerland, this species
also had a univoltine life cycle (Breitenmoser-Wursten and Sartori, 1995).
Wise (1980) also found that the temporal distribution and development of
Rhithrogena semicolorata typically a ‘winter’ species was apparently
limited by the warmer temperatures which prevailed for a longer period in
the lower reaches during the summer. Thus, in these lower reaches,
emergence was sooner and hatching took place later than in the hill region.
Hence, the period during mid-summer when the nymphs were absent from
the benthos was of correspondingly longer duration. This pattern was also
observed during the present study. For example, in Callow Loughs Stream,
Rhithrogena semicolorata was absent in July, August and September 2001
and hatched in October into young larvae. It emerged predominantly in
March and April 2002 with a few remaining in May. The nymphs were
198
absent for a period of three months and began hatching into immature larvae
in October 2002. The situation was slightly different in the Dunneill River.
Young larvae were only absent in July 2001 and began to hatch into the
system in August to commence a new generation. It began to emerge in
March 2002 and continued until the end of June. Immature larvae were
absent for a shorter period (two months) in this river and appeared again in
the system in August 2002. Interestingly, the larvae in the Dunneill River
grew to larger sizes (12-13mm) prior to emergence in comparison to the
smaller sized larvae (9-10mm) found in Callow Loughs Stream at the onset
of the emergence period. It is reasonable to suggest, therefore, that this was
due to the larvae spending a shorter period of time (7-8 months) growing in
the benthos, particularly during the winter period when temperatures were
low, producing smaller sized animals. The larvae in the Dunneill River
grew to larger sizes presumably as a consequence to spending a longer
period of time (8-9 months) in the benthos reaching body lengths of
between 12-13mm.
199
4.5.4 Interpretation of the life history of Heptagenia species
Heptagenia spp. was found in the five highs status sites but due to
insufficient numbers captured during the sampling period in the Brusna and
the Dunneill Rivers the life cycle was not interpreted in these sites. It was
the least abundant of all the Heptageniidae species studied in this
investigation. The specimens were identified to genus level only so it is
possible that a number of species were among this genus. The Heptagenia
specimens were therefore described as a genus group and appeared to adopt
a univoltine life cycle with an overwintering larval generation.
Wise (1980) noted that Heptagenia lateralis (Curtis) was also univoltine
with an overwintering generation that grew rapidly prior to emergence in
early summer. He noted overall that Heptagenia lateralis grew little during
the winter and emerged later than Rhithrogena semicolorata after a period
of rapid spring growth. Rhithrogena semicolorata was absent from the
benthos for a period during the summer between the completion of
emergence and the beginning of the next generation.
Heptagenia spp. was poorly represented in the Dunneill River, Callow
Loughs Stream and in the Brusna River. The life history of this genus
displayed very similar patterns in the Owengarve and the Castlebar River.
In both, larvae were absent from the benthos in July and August 2001 and
hatched in September 2001. During the autumn and winter small instars
were present but growth was minimal. In March and April, an outburst of
hatching activity coincided with an increase in growth rate followed by its
emergence in June 2002. A new generation hatched a month earlier in the
Owengarve River (August 2002) compared to the Castlebar River where the
next generation hatched in September 2002. Recruitment continued on into
October 2002 as juvenile larvae increased in numbers. It is worth noting
that in all sites, Heptagenia spp. emerged later than Rhithrogena
semicolorata
200
4.5.5 Rearing trials
In general, the success of the rearing trials was quite low. This was
especially true for the successful emergence of Ecdyonurus venosus from
nymph to adult. Quite a large number of larvae of this species were
identified in the rivers but failed to survive beyond a few days after
transferring to the aquarium. Interestingly, larvae of Ecdyonurus insignis
and Ecdyonurus dispar had better survival rates and emerged more
successfully than Ecdyonurus venosus. This species may have been more
sensitive to the type of algal slime that grew in the tank causing death.
Alternatively, the copper pipe used to connect the reservoir to the aquarium
tank may have caused a small degree of toxicity that this species is more
sensitive to compared to the others, as all three species were reared at the
same time and exposed to the same water. These hypotheses however
remain purely speculative.
Finlay (2001) found greater rearing success of mayfly larvae in small
individual chambers with much less water (1.25 litre plastic soft drink bottle
cut in two and oxygenated) compared to large flow tanks powered by
propellers. He considered this to be indicative of an inherent need for
highly oxygenated water in some genera. For instance, a high rate of water
movement may be necessary for the development of Ecdyonurus venosus
that may not have been adequate in the aquarium used in our rearing
experiments. It is possible that it requires a more concentrated supply of
well oxygenated water to survive. It also indicates, that Ecdyonurus insignis
and Ecdyonurus dispar may need lower current flow to complete its life
cycle successfully. As mentioned previously, these rearing trials were
designed solely for the verification of the various species of Ecdyonurus,
particularly where a few species were present at the same time in the same
river.
201
It was hoped that they would also help in confirming emergence periods
among this genus and that the timing would be similar to those described
from the studies carried out over 2001 and 2002. Due to high mortality
rates however, it was difficult to confirm the exact timing of the emergence
of the different species of Ecdyonurus in the rearing experiments. Rearing
the nymphs to the adult stage did however assist in verifying specimens
especially as in the case in the Owengarve River, where three species were
present in the river during the summer months. It must be noted also, that
due to the warmer temperatures of the aquarium water, the larvae tended to
emerge earlier in comparison to how they might do in their natural state in
rivers.
4.5.6 Concluding comments
In order to ensure that the Heptageniidae, in particular the genus
Ecdyonurus are fully represented when carrying out routine biological
assessments, the timing of sampling may be a critical factor, especially
when attempting to capture species that are only present in the benthos for a
few months during the year. Findings from the present studies appear to
show that more intense sampling may be required during certain months of
the summer. This would ensure a complete characterisation of the range of
species present in a river system that may not be there if the sampling were
carried out during the autumn/winter months.
From studying the life cycles of the Heptageniidae in five rivers in the West
of Ireland it would appear that the life history of the genus Ecdyonurus can
vary slightly from year to year and from river to river. Due to the short life
cycle of Ecdyonurus insignis and in particular Ecdyonurus dispar, which are
summer species only, it would be vital to sample from June to August in a
given year to ensure capture of these species. The more common species of
this genus, Ecdyonurus venosus was present in each of the rivers studied
throughout the year, with the exception of a few months post emergence.
Hence, one would expect to find this species when sampling during all
seasons throughout the year. Apart from being absent from the benthos for
a few months after the main flight period, particularly in July and August,
202
Rhithrogena semicolorata was also present throughout the year, so one
would therefore expect to find this species from September to June also.
Hepatgenia spp. emerged a bit later than Rhithrogena semicolorata so
depending on the abundance in a river, would be expected to be found
throughout the year when routinely sampling, apart from July and August
and in some rivers in September.
As mentioned previously, positive identification of species requires
examination of all life stages for most aquatic insects. The two main
approaches are field and laboratory rearing. An attempt was made in this
study to collect both nymphs (rearing trials in the laboratory) and adults
(field studies) from each location to verify individual species.
Unfortunately, these studies were unsuccessful due to high larval mortality
in the laboratory aquarium and also to the difficulty in finding adults in the
field caused by time restrictions that went beyond the constraints of the
project. In addition, this study did not include an investigation into whether
the eggs and larvae overwintered and as mentioned previously, assumptions
that they did, were based on findings from Elliott and Humpesch (1983),
Macan and Maudsley (1968) and Wise (1980). The interpretation of water
quality assessment programmes may be limited due to the knowledge gaps
that still remain therefore further studies in these particular areas are
essential to complete a more accurate and detailed description of the life
cycle of the Heptageniidae in the West of Ireland.
203
Chapter 5
Assessment of the physical, chemical and biological factors
controlling the occurrence of Ecdyonurus in five high status
and five impacted rivers in the West of Ireland.
5.1 Introduction
As previously mentioned in Chapter 4, Ecdyonurus is a key water quality
indicator genus. Its sensitivity to water pollution is noted in a wide range of
biological indicator systems, based on macroinvertebrates as are most
members of the Heptageniidae (e.g. De Pauw and Vanhooren 1983; Moog
1995; Skriver et al., 2001). Ecdyonurus, a widely distributed genus (Kelly-
Quinn and Bracken, 2000) is particularly important in Ireland for the Irish
EPA’s Quality Rating System (Q-value) (McGarrigle 1998; McGarrigle et
al., 1998; McGarrigle pers. comm.). The Quality Rating System in turn
shows clear statistical links between Q-values and water quality parameters
such as BOD, ammonia, nitrate and phosphate, all being important chemical
indicators of water quality (Clabby et al., 1992; McGarrigle et al., 1998).
The Irish Phosphorus Regulations (DELG 1998) are based on a link
between biological Q-values and unfiltered molybdate reactive phosphorus
(MRP) concentrations. While these strong empirical links exist, the precise
mechanisms controlling the distribution of Ecdyonurus in eutrophic and
polluted rivers are not well understood or documented. Ecdyonurus is one
of the most useful indicators of pollution because it is ubiquitous in
unpolluted waters but sensitive to pollution.
The occurrence of Ecdyonurus is probably not controlled by one single
factor. It is likely that three aspects of the river environment control the
presence of this sensitive indicator i.e. chemical, physical and biotic factors.
It is therefore necessary to integrate all three areas in order to provide an
overall understanding of how Ecdyonurus and other sensitive species
respond to pressures on their environment.
Monitoring the quality of water in a stream as an indicator of catchment
health should include biological response attributes as well as an assessment
204
of the physico-chemical characteristics. Changes in the health of a system
will be reflected in the aquatic biological community if they are exposed to
environmental disturbance or stress as they act as integrators of the multiple
present and past environmental effects (Cranston et al., 1996).
Bioindicators at the species level rather than higher taxonomic groups,
appear to be more useful in detecting environmental stress due to the
sensitivity that individual species have to environmental factors (Moog,
1995). The use of the indicator approach in aquatic ecosystems, in
particular the use of macroinvertebrates, has received considerable
attentions in biomonitoring programs (Rosenberg and Resh, 1993; Johnson,
1995). Biomonitoring of communities with an emphasis on taxonomic
richness and composition is considered by many advocates to be a most
sensitive means of detecting alterations in aquatic ecosystems. However, an
even earlier warning of stress may be provided by physiological and
morphological abnormalities in some aquatic biota (Clarke, 1994;
Rosenberg and Resh, 1993; Madden et al., 1995; Clarke et al., 1995).
Whole catchment responses are well reflected in the aquatic system because
the water flowing from a catchment actually does the integration.
Parameters measured at the bottom of the catchment integrate what goes on
in the catchment per se. Thus, information obtained at a single monitoring
site per catchment at its bottom-most point may summarise the total
catchment response (Cranston et al., 1996). Obviously, however, questions
of scale and extent of a particular impact also have to be considered
especially when dealing with large catchments. There is also a suggestion
that the true impacts on first order streams, which may be important for
spawning of trout, for example, may not be adequately represented by
sampling larger downstream streams (McGarrigle pers. comm.).
The primary physical characteristics or “key factors” which influence
aquatic organisms are temperature, oxygen content, nutrient composition
and availability and habitat structure or cover (Moog, 1995). Temperature
has an important effect on the intensity of metabolic activity as well as most
other biological processes of aquatic organisms. Temperature is also of
decisive importance for the occurrence of specific organisms in an
205
environment and hence can determine the composition of a community.
While mean temperatures and temperature sums (degree days) are of some
importance (especially in determining growth rates and time of emergence
for insects), it is primarily the temperature range or extreme values which
determine the suitability of a site as habitat for a particular species. Both the
optimal and the tolerable temperature ranges for some benthic organisms or
their specific life stages e.g., egg, larvae, pupa and adult, are already known
and can be used in assessing some temperature problems (Moog, 1995).
The response of macroinvertebrates to oxygen concentration is relatively
unidirectional as their distribution is never limited by increasing oxygen
saturation and only rarely is oxygen supersaturation considered damaging to
organisms. Therefore, evaluating oxygen-dependent distributions is limited
to determining an organism’s sensitivity to low levels of oxygen particularly
in regards to their ability to recover from extremely low oxygen
concentrations. Organisms that are most sensitive to oxygen deficiencies
are those with thick skins and no gills (many Plecoptera larvae) or
immovable gills (Ephemeroptera larvae of the genera Epeorus, Rhithrogena
etc.). Although oxygen content alone is insufficient in describing the
saprobic conditions of a river it is nevertheless an important factor in
discriminating water-quality classes (Moog, 1995). In eutrophic systems,
supersaturated oxygen values measured in daylight may be regarded as a
sign of potentially low oxygen saturation values at night time as the plant
biomass which drives supersaturation by photosynthesis may be sufficient to
cause oxygen depletion at night due to respiration (McGarrigle 2001). The
Irish EPA regards supsersaturation values of over 120% as an indication of
eutrophication in rivers and potential nocturnal deoxygenation (McGarrigle
et al., 2002).
Nutrients and their availability are key factors in controlling trophic-
relations within a river system. Seasonal changes in nutrients inputs causing
algal blooms or even the sudden organic input of leaf-fall in autumn can
cause changes in the abundance of benthic macroinvertebrates, in particular
to sensitive indicator taxa such as the genus Ecdyonurus (McGarrigle et al.,
1998).
206
Factors controlling the distribution and abundance of benthic invertebrates
have been the subject of considerable debate (Stanford and Ward, 1983;
Lake and Barmuta, 1986; Townsend, 1989; Townsend and Hildrew, 1994).
The habitat or substrate structure requirements of benthic species is very
important in determining their distribution within a river system. It has been
suggested that there may be an interaction between areas of stream habitat
that show variations in substrate content or that have been disturbed
differently and invertebrate richness found in a stream (Lake and Barmuta,
1986; Townsend and Hildrew, 1994). Additionally, unexplained variability
in the spatial distribution and abundance of stream invertebrates is the likely
basis for the requirement to collect a large numbers of replicates to be
representative of a site (Chutter, 1972; Schwenneker and Hellenthal, 1984;
Norris et al., 1992), particularly when a Surber sampler is used.
Stream habitat forms an essential component of river ‘health’ (Maddock,
1999) that can be used to evaluate the overall ecological integrity of a river
system (Muhar and Jungwirth, 1998). The condition of local stream habitat,
otherwise known as the habitat template, influences the structure and
organisation of biological communities (Hynes, 1970; Southwood, 1977;
Swanson, 1980; Minshall, 1984; Sweeney, 1984; Downes et al., 1995). In
the absence of water quality impairment, the local physical habitat will have
a major influence over the biotic assemblages at a site. If water quality is of
a high standard, diverse and abundant assemblages of stream biota are likely
to exist if the local stream habitat is supportive (Plafkin et al., 1989;
Barbour et al., 1999; Simpson and Norris, 2000).
Contemporary assessment of river health often involves some form of rapid
biological assessment (Reynoldson et al., 1997; Norris and Thoms, 1999).
Rapid assessment approaches have focused on aquatic organisms.
Empirical models have been developed that predict the occurrence of
macroinvertebrate taxa, based on their association with environmental
variables at reference sites (Wright, 1995; Reynoldson et al., 1997; Simpson
and Norris, 2000). This approach provides an independent way of matching
new sites with reference sites, enabling predictions to be made. Thus, the
207
philosophy of this approach seems appropriate. However, these techniques
have not been applied to features of streams other than their biota. Stream
managers often require information about physical features of a river that
need improving to enhance biological condition. Most of the rapid
assessment approaches have limited ability to determine whether biological
impairment results from poor water quality or from poor habitat. Therefore,
the ability to predict local stream habitat features may be useful for
distinguishing between the effects of water quality and the effects of habitat
on biological condition, and may assist in river management (Davies et al.,
2000).
Before a habitat can be identified as damaged, it is vital to know what the
habitat should be like in the absence of effects from humans. Stream habitat
may be influenced by a variety of factors operating at numerous spatial and
temporal scales (Frissel et al., 1986; Richards et al., 1996). Habitat
assessment is beginning to focus on nationally applicable and ecologically
based approaches that incorporate comparisons to reference or target
conditions. The River Habitat Survey (Raven et al., 1997, 1998; Jeffers,
1998) and the United States Environment Protection Agency Rapid
Bioassessment Protocols (Plafkin et al., 1989; Barbour et al., 1999) use
reference conditions to assess stream habitat at broad scales (state-wide or
nationally).
208
5.2 Study outline
This study focuses on obtaining information on the value of the physico-
chemical data in supporting the Q-value System and compares the physical,
chemical and biological data as indicators of water quality between five
high status and five impacted rivers studies in the West of Ireland.
209
5.3 Materials and Methods
Details of the ten river sites studied in this investigation are outlined in
Chapter 1.
5.3.1 Preliminary site investigations
In order to establish the suitability of a river site to the project requirements,
biological and chemical investigations were carried out in a number of
rivers. A kick sample was taken in a total of 29 rivers and biologically
assessed by assigning a Q-value to each river. The multihabitat kick sample
was collected in each river using a 0.25 x 0.25m sweep net with a 500µm
mesh. The net was trawled with an ‘S’ type movement behind feet kicking
the substrate for three minutes. Where possible, all available habitats were
sampled by kicking, stone washing and weed sweeping.
To assess the water quality of each river, water samples were analysed for
the following physico-chemical parameters: Temperature (taken in the
field), dissolved oxygen % saturation (taken in the field), pH, conductivity,
ortho-phosphate, chloride, BOD, ammonia, Total Oxidised Nitrogen
(T.O.N.), alkalinity and colour. Analytical methods for the water quality
parameters are outlined in Table 5.1.
High status rivers which are potentially close to their reference condition
were chosen on the basis of a high Q-value, the presence of the genus
Ecdyonurus and good water chemistry. Impacted sites chosen had lower Q-
values (3 or 3-4) where at one time Ecdyonurus survived but are now no
longer present. Five high status and five impacted rivers were then chosen
for this project. The list of high status rivers and impacted rivers selected
for this study and associated information is shown in Table 1.1 and Table
1.2 outlined in chapter 1.
210
Water samples were collected in a 2-litre HDPE water-sampling bottle filled
from below the surface of the water. All sampling bottles were washed
thoroughly with water from each site before taking the final sample.
Samples were transferred to the laboratory refrigerator (5°C) on the same
day and analysed within 24 hours of sampling.
A summary of the methodologies used in analysing the water quality
parameters listed above is outlined in Table 5.1.
Table 5.1 Summary of the methodologies used in analysing the water
chemistry parameters.
Parameter Methodology
Detection
range
Unit
Temperature
WTW DO meter °C
Dissolved Oxygen
WTW DO meter mg/l O2
pH WTW inolab terminal
level 3 meter
2-12 pH units
Conductivity WTW inolab terminal
level 3 meter
15-12,890 µS/cm 25oC
Alkalinity Titration using sulphuric
acid
8-150 mg/l CaCO3
Colour DR4000 Hach
Spectrophotometer
5-500 Hazen units
Ammonia KONELAB 30
autoanalyser
0.03-1 mg/l N
TON KONELAB 30
autoanalyser
0.4-10 mg/l N
Orthophosphorus KONELAB 30
autoanalyser
0.012-0.5 mg/l P
Chloride KONELAB 30
autoanalyser
2-100 mg/l Cl
BOD
5 day incubation
@ 20°C
1-7 mg/l O2
211
5.3.2 Physico-chemical parameters - procedures and analysis
Water samples were analysed in the ten rivers on a monthly basis for the
chemical ‘quality’ parameters outlined in Table 5.1. Sampling began in July
and August 2001 depending on the river and continued on a monthly basis
until October 2002. The sampling programme took place over a period of
16-18 months depending on the river. Monthly samples were generally
taken in all of the high status sites. However, due to adverse weather
conditions some of the rivers were not sampled in January 2002. All
sampling was completed halted in February because of flooding (chemistry
samples were normally taken in conjunction with macroinvertebrate
samples). Water samples were taken from the impacted sites from July to
October 2001 and again from March to October 2002.
212
5.3.3 Macroinvertebrate collection
5.3.3.1 Sampling procedure
Quantitative macroinvertebrate sampling commenced in July 2001 in four
(Owengarve River, Castlebar River, Dunneill River and Callow Loughs
Stream) of the five high status rivers. Sampling began a month later in the
Brusna River during August 2001. Monthly sampling continued for 16
months until October 2001. From November 2001 to January 2002
inclusive, kick samples were taken, as the water levels were too high to
carry out routine Surber sampling. Due to floods in February 2002,
sampling was completely suspended. Monthly Surber sampling
recommenced in March and continued until October 2002. The sampling
programme took place over a period of 16 months. These macroinvertebrate
samples were also used to study the life cycle of the genus Ecdyonurus
described in Chapter 4.
In three (the Mullaghanoe River, Lough na Corralea Stream and the Cartron
River) of the five impacted rivers, monthly Surber sampling was carried out
from July 2001 to October 2001 inclusive and again from March 2002 until
the sampling programme ceased in October 2002. No sampling was carried
out in these rivers from November 2001 to February 2002 inclusive (it was
felt better to extend the seasonal range of the study into late 2002 than to
cease three months earlier) – logistics dictated the total number of samples
that could be sorted and identified within the project timescale. Surber
sampling commenced earlier (May 2001) in two of the impacted sites,
namely the Robe River and Lough na Corralea Stream. The Robe River and
the Mullaghanoe River both contained specimens of Ecdyonurus on
occasions. Depending on the water levels, it was decided therefore, to take
a kick sample in these rivers during the winter months to obtain Ecdyonurus
specimens. No further sampling was undertaken during the winter period.
The experimental design and sampling regime are described in section 4.3.1
of Chapter 4 and the sampling areas for each river have been detailed in
Chapter 1. Within each quadrat the Surber was also placed randomly and
the macroinvertebrate sample taken. Five Surbers were taken in each river
213
on each date. The Surber sampler covered an area of quantitative
macroinvertebrate sampling in each river site and was carried out using a
1/16 m
2 horizontal base (25cm x 25cm) and had an attached pond net with
mesh of 650µm. The net was held open by a square-foot metal frame 36cm
x 28cm. In operation, the frame that supports the net is in a vertical
position, while the other frame is locked into a horizontal position against
the bottom. The net opening was placed facing upstream, using the current
to hold the net open and the horizontal frame was pushed into the river
bottom (Standard Methods, 2001). The macroinvertebrates were collected
from the riverbed by disturbing the substrate within the metal frame for 2
minutes and washing all stones thoroughly to dislodge animals. The large
substrate material was removed and the remaining sample was transferred
into a 1L plastic container and preserved immediately in 70% Industrial
Methylated Spirits (IMS).
214
5.3.3.2 Determination of optimum numbers of Surber samples
A large variation is usually encountered in sampling natural
macroinvertebrate populations with the result that small samples or small
numbers of samples can be statistically inaccurate and imprecise.
Therefore, when sampling quantitatively, it is vital that an adequate number
of sampling units is taken to ensure an optimum level of accuracy in the
sense of obtaining a mean population value that is as close as possible to the
true mean for the population in question. In order to establish the optimum
number of Surber samples for each river site, that maximises the return on
effort, a series of samples were taken from a number of sites. Elliott (1979)
suggests taking 5 sampling units at random and calculating the arithmetic
mean. Next take 5 more units at random and calculate the mean for 10
units. He advises to continue to increase the sample size by 5-unit steps and
plot the means for 5, 10, 15 etc. units against sample size. When the mean
ceases to fluctuate, a suitable sample size has been reached and this sample
size can be used for that particular station.
This process was carried out in one high status river (Dunneill River) and
one impacted river (the Mad River). A total of 20 Surbers were taken in
each site. On the basis of tolerating a target of 20% of the mean (as
suggested by Elliott, 1979) the analysis showed that five Surbers was a
suitable sample size to be taken from each site. For two of the rivers,
increasing the number of Surber samples beyond this resulted in only a
small improvement in accuracy and thus significant diminution of return on
effort. Results of the percentage recovery of macroinvertebrates in 20
Surbers from the Dunneill River on 4th
July 2001 are outlined in Fig. 5.1.
215
Fig. 5.1 Percentage recovery of macroinvertebrates collected from 20
Surbers in the Dunneill River on 4th
July 2001.
Graphing the cumulative frequency of the presence/absence data showed,
however, that a total of ten Surbers needed to be taken from the impacted
site (Mad River) due to the high variance recorded there. Fig. 5.2 outlines
the percentage recovery of the macroinvertebrates from the Mad River taken
on 21st July 2001.
Fig. 5.2 Percentage recovery of macroinvertebrates collected from 20
Surbers in the Mad River on 31st July 2001.
0
20
40
60
80
100
1 3 5 7 9 11 13 15 17 19
Surber number%
R
eco
ver
y
0
20
40
60
80
100
1 3 5 7 9 11 13 15 17 19
Surber number
% R
eco
ver
y
216
5.3.3.2 Analysis of sample data 5.3.3.3
All macroinvertebrate Surber samples taken from the high status or potential
reference condition rivers during the study programme were sorted entirely
– i.e. five samples per month. Due to time constraints it was not feasible to
fully examine all the Surber samples from the impacted sites. It was
decided to concentrate firstly on the Robe and the Mullaghanoe Rivers as
some of the Surber samples taken from these rivers contained Ecdyonurus
on occasion. As many Surber samples as possible were sorted from the
three remaining rivers within the time available thereafter. Table 5.2
outlines the number of Surbers that were sorted and identified for each of
the impacted rivers investigated.
Table 5.2 The number of Surbers sorted per month in the impacted rivers
during the study programme.
Month\River Mullaghanoe Robe Mad
Cartron Lough na
Corralea Stream
May 2001 * 5 * * 5
June 2001 * * * * *
July 2001 5 * 20 4 5
August 2001 4 2 2 4 5
September 2001 3 2 * 2 *
October 2001 1 * * 3 *
March 2002 1 2 * 1 *
April 2002 5 * * 5 *
May 2002 * 2 * 5 5
June 2002 4 2 * * 5
July 2002 4 * 1 5 3
August 2002 4 * 1 5 2
September 2002 3 2 * * 1
October 2002 3 2 * * *
* = No Surbers were sorted for this month
Total number of taxa (S), total number of individuals (N) (abundance) and
percentage Ephemeroptera, Plecoptera and Trichoptera (% EPT) were
complied for all sites and dates. Four indices were also calculated. These
included included the Shannon-Wiener index (H’), Simpson index (λ),
Margalef’s index (d) and Pielous’ evenness index. These metrics were used
to describe the community structure and highlight any differences between
the high status and impacted rivers examined in this study. Different
diversity indices emphasise the species richness or equability components of
217
diversity to varying degrees. The most commonly used diversity measure is
the Shannon (or Shannon-Wiener) diversity and the Simpson indices. These
two non-parametric indices were used to test for heterogeneity.
The Shannon-Wiener index (H’) defined as (Krebs, 1989)
H’ = - ∑i pi log (pi)
where pi is the proportion of the total count (or biomass etc.) arising from
the ith species. H=0 if there is only one species in the sample and is
maximum when all species are represented by the same number of
individual (even distribution of abundance; Green, 1979; Ludwig &
Reynolds, 1988).
The Simpson index (λ) defined as
λ = ∑ pi2
1-λ = 1 – (∑ pi2)
λ’ = {∑i Ni (Ni –1)}/{N(N-1)}
1-λ’ = 1 – {∑i Ni (Ni –1)}/{N(N-1)}
where Ni is the number of individuals of species i. The index λ has a natural
interpretation as the probability that any two individuals from the sample,
chosen at random, are from the same species (λ is always ≤ 1). It is a
dominance index, in the sense that its largest values correspond to
assemblages whose total abundance is dominated by one, or a very few, of
the species present. Its complement, 1-λ, is thus an equitability or evenness
index, taking its largest value (of 1 – S’) when all species have the same
abundance. The slightly revised forms λ’ and 1-λ’ are appropriate when
total sample size (N) is small.
Taxon richness is often given simply as the total number of taxa (S), which
is obviously dependent on sample size (the bigger the sample, the more
species there are likely to be). Alternatively, richness can be assessed using
218
Margalef’s index (d), which incorporates the total number of individuals (N)
and is a measure of the number of taxa present for a given number of
individuals:
d = (S-1) / log N
Equitability expresses how evenly the individuals are distributed among the
different species, and is often termed evenness. It is expressed as Pielous’
evenness index:
J’ = H’ / H’max = H’ / log S
where H’max is the maximum possible value of Shannon diversity i.e. that
which would be achieved if all species were equally abundant (namely, log
S).
The box and whisker plots included in Section 5.4.3.1 to 5.4.3.7 show the
variations in the diversity indices and biotic metrics on a river by river basis
across the entire sampling period. The various diversity indices emphasise
the species richness or equitability components of diversity to varying
degrees. The Shannon-Wiener index reflects an increasing diversity with a
higher index number. For example, a value of 3 indicates a higher diversity
than a value of 0. Simpson’s index is similar to the Shannon-Wiener index
and is a measure that accounts for both richness and proportion (percent) of
each species. Margalef’s index reflects the overall species richness within a
community where increased species richness increases with a higher index
number. Pielou’s evenness is a component of diversity and compares and
quantifies faunal equitability to taxa diversity for a given area. In general,
all of the diversity indices in this study showed a similar trend across the
study sites, with a higher diversity being found in the high status rivers
compared to the impacted rivers (see results section).
The diversity indices described above were used to reduce the multivariate
(multi-species) complexity of assemblage data into a single index (or small
number of indices) which was then handled statistically by univariate
analysis. ANOVA was run on the derived indices and metrics to test for
differences in the community structure between the high status and
219
impacted rivers. It was decided to omit the data obtained during the winter
months (November 2001, December 2001 and January 2002) from these
analyses as they represented kick samples only and not Surber samples. It
was thought that the data obtained from these months would be inconsistent
with that compiled during the rest of the study.
In addition to the above indices, a set of metrics and indices to assess quality
in lotic ecosystems (see Brabec et al., 2004; Buffagni et al., 2004; Ofenbock
et al., 2004) were computed for both the high status and the impacted sites
over the study period (Table 5.3). These were calculated using the recently
available AQEM project software package (AQEM Version 2.3.4a, 2004).
The indices were chosen based on their suitability for assessing the impact
of organic pollution and eutrophication.
Table 5.3 Definition of selected indices/metrics.
Index/metric Definition Symbol Reference
Belgian Biotic Index Combination of richness
a tolerance of selected
BBI De Paw and Vanhooren, 1983
Danish Stream Fauna
Index
Combination of richness
a tolerance of selected
DSFI Skriver et al., 2001
*German Saprobic Index Indication or measure of
the level of organic
pollution
GERMAN DEV, 1987, 1992; Pantle and Buck, 1955
**Czech Saprobic Index Indication or measure of
the level of organic
pollution
CZECH Sladecek, 1973; Rotschein, 1982
***Dutch Saprobic Index
Indication or measure of
the level of organic
pollution
DUTCH Zelinka and Marvan, 1961
Indice Biotico Esteso
Sum of selected tolerance
taxa
IBE Ghetti, 1997
Average Score per Taxa ****BMWP divided by
the existent selected taxa
ASTP Alba Tercedor and Sanchez Ortega, 1988
Saprobic Index (Zelinka and Marvan, 1961) = SI z&m = ∑ s
z&m si.
s z&m gi. n I
∑ s
z&m gi.ni
SI Z&M : Weighting factor (szg)
** The Czech Saprobic Index is calculated exactly the same way as the Saprobic Index (Zelinka and Marvan,
1961) including the weighting factor but with a slightly different taxa list.
*** The Dutch Saprobic Index is calculated in exactly the same way as the Saprobic Index (Zelinka and Marvan,
1961) but without the weighting factor.
****BMWP – Certain macroinvertebrate families are scored according to their sensitivity to organic pollution.
The BMWP is the total of the scores of all families present in a taxa list. Each family in the sample is counted
only one time, regardless of the number of species.
220
The objective of calculating the indices in Table 5.3 was to evaluate the
robustness of the assessment methodologies to discriminate between the
contrasting group of quality status rivers (high status and impacted sites).
These metrics are commonly used to assess organic pollution. The Danish
Stream Fauna Index (DSFI) (Skriver et al., 2001) have been also used to
assess general degradation (AQEM consortium, 2002). For some of the
metrics and indices, boundaries between quality classes are already
established. These boundaries are referred to in Table 5.4. Interpretation of
the boundaries for the remaining indices that are not outlined to in Table 5.8
were taken from the output data in the software.
Table 5.4 Class boundaries used as scoring criteria for some of the studied
metrics.
Boundary
Metric
High/Good Good/Moderate Moderate/Poor Poor/Bad
BMWP 100 60 30 15
ASPT 0.50 0.43 0.34 0.25
BBI 8 6 4 2
IBE 8 6 4 2
GERMAN 1.5-2.2 2.2-3.0 3.0-3.5 >3.5
CZECH 1.5-2.2 2.2-3.0 3.0-3.5 >3.5
The AQEM software delivers results at different levels, which can be used
to specify management implications and procedures. The AQEM
assessment method is based on a “multimetric” procedure. A multimetric
index combines several individual formulas (e.g. saprobic indices, feeding
type composition, microhabitat etc.), the results of which are finally
combined into a multimetric result. Thus, multimetric indices integrate
multiple attributes of stream communities (“metrics”) to describe and
evaluate a sites condition (AQEM consortium, 2002).
One-way ANOVA was then carried out on the indices/metrics, microhabitat
preferences and the feeding types derived from the AQEM software to
assess the differences between the high status and the impacted sites.
221
5.3.4 Sediment analysis
As part of the overall habitat assessment, substrate sampling was carried out
in all ten rivers between the 21st and the 27
th March 2003. Sampling and
analysis of the river bed material was carried out in accordance with the
Central Fisheries Board standard methodology for sediment analysis (King
and Kelly, 1999).
5.3.4.1 Sampling procedure
Sampling was carried out in an area of uniform appearance with regard to
bed material. A metal-framed Surber sampler with a horizontal metal frame
measuring 1/16 m
2 (25cm x 25cm),
with a 650µm pond net attached was
placed on the stream bed as described in the previous section. A fine
plankton (35µm) net with screw off bottle at end was placed over the
coarser Surber net to retain fine silt material. Using a spade to dig vertically
to loosen material inside the quadrat, particles were removed directly into a
10L metal bucket or sampling bag. Fine material worked backwards into
the fine mesh net in a downstream manner after passing through the Surber
net. Material was collected to a depth of approximately 0.3m. Three
random Surber samples were taken in each river across a transect as
follows: close to the right bank, central channel area and close to the left
bank.
5.3.4.2 Sample Treatment
The substrates were allowed to air dry for 4 weeks prior to analysis. The
entire sample was hand sieved using a 256mm-aperture sieve to remove the
larger fraction. The sediments were then sieved manually through 4 grades
of steel sieves (64mm, 32mm, 16mm, 8mm). All remaining material was
dried completely and sieved for 20 minutes through a stack of sieves
(aperture 128 to 0.03mm) using an intermittent amplitude mode capable of
sorting the coarsest fractions of the sample. Each fraction retained in each
sieve was weighed to the nearest gram and recorded.
222
5.3.4.3 Analysis of samples
The relationship between sieve aperture (mm) and the commonly used
descriptive scale American Geophysical Union Descriptive Scale
(A.G.U.D.S.) for substrate analysis is shown in Table 5.5. Particle size is
frequently expressed on a logarithmic scale in which each successive size
fraction covers twice the range of the proceeding one, e.g. 0.25, 0.5, 1, 2,
4mm etc. or for each phi unit increase in size, the corresponding particle
size in millimetres is halved and for each phi-unit decrease in size the
corresponding particle size is doubled (Table 5.5).
Table 5.5 Phi-unit to millimetre and log-scale conversion table and the size
ranges of the particle size categories and sub-categories. Adapted from
Gordon et al. (1992).
A.G.U.D.S Sieve Aperture
(mm)
Phi units Log scale
Boulders 256-512 -9 2.7
Large cobble 128-256 -8 2.4
Small cobble 64-128 -7 2.1
Very coarse gravel 32-64 -6 1.8
Coarse gravel 16-32 -5 1.5
Medium gravel 8-16 -4 1.2
Fine gravel 4-8 -3 0.9
Very fine gravel 2-4 -2 0.6
Very coarse sand 1-2 -1 0.3
Coarse sand 0.5-1 0 0
Medium sand 0.25-0.5 1 -0.3
Fine sand 0.125-0.25 2 -0.6
Very fine sand 0.063-0.125 3 -0.9
Coarse silt 0.032-0.063 4 -1.2
223
5.3.4.4 Data presentation
The percentage composition by weight (g) of each fraction from each
sample was calculated. The arithmetic mean percentage composition of
each fraction was also calculated.
Cumulative percentage frequency curves were plotted for each of the three
samples together with the calculated arithmetic mean. Log 10 sieve aperture
(mm) was employed on the x-axis, as sediment particle sizes tend to follow
a logarithmic distribution.
High status and impacted sites were all analysed individually. Comparison
of fines content (sediment particles < 1mm) in samples was analysed using
ANOVA to identify any differences between the individual sites and to
highlight any differences between the high status and impacted sites. The
relationship in the mean percentage frequency of all substrate fractions
between the high status and impacted sites were also analysed using
ANOVA.
5.3.5 Night-time Dissolved Oxygen (DO) measurements
Eutrophication has a major impact on dissolved oxygen levels in river
water. Excessive plant growth can cause significant diel fluctuations in
dissolved oxygen levels in rivers (Moriarity, 1990; McGarrigle 2001). In
order to establish whether minimum dissolved oxygen saturation at night
was an important factor controlling the occurrence of Ecdyonurus, it was
essential to carryout night-time DO measurements in all ten rivers. This
was undertaken on the 7th
and 14th
August 2003 at a time of low flows and
high water temperatures. Single measurements for both DO and
temperature was taken in each river using a WTW (Wissenchaftlich-
Technische Werkstatten) oxygen probe during the hours of darkness.
224
5.4 Results
5.4.1 Physico-chemical results
The maximum, minimum, mean, median and standard deviation values for
the water chemistry parameters examined in the high status and impacted
sites during the sampling programme are outlined in Appendices 5.1 and 5.2
respectively. Rainfall data for the 2001-2002 period is shown in Appendix
5.3 as an aid to the interpretation of the water chemistry data and in order to
give an idea of the potential flow variation in the rivers studied.
Graphical illustration of the mean monthly variations are presented for
temperature (Fig 5.3) and pH (Fig 5.4), conductivity (Fig 5.5) and alkalinity
(Fig 5.6), dissolved oxygen (DO; Fig 5.8) and biochemical oxygen demand
(BOD; Fig 5.9), colour (Fig. 5.10) and chloride levels (Fig 5.11), ammonia
(Fig 5.12), total oxidised nitrogen (TON; Fig 5.13) and unfiltered molybdate
reactive phosphorus (MRP; Fig 5.14). Sampling was not carried out during
some of the winter months due to bad weather thereby creating the gaps in
the dataset shown in some of the graphs.
225
5.4.1.1 Temperature (°°°°C) and pH
The temperatures observed in the high status and impacted sites during the
sampling period were within the normal seasonal ranges (Fig. 5.3). The pH
values were mostly circum-neutral or alkaline but were slightly lower on
occasion in three of the impacted sites (Cartron River, Lough na Corralea
Stream and the Mad River) compared to all other rivers investigated in this
study reflecting the different geology and typology in the individual
catchments (Fig. 5.4)
Fig. 5.3 Variation in mean temperature values in samples of stream water
collected on a monthly basis from the high status and impacted rivers during
the study.
Fig. 5.4 Variation in mean pH values in samples of stream water collected
on a monthly basis from the high status and impacted rivers during the
study.
5.4.1.2 Conductivity (µµµµS/cm) and alkalinity (mg/l CaCO3)
0
2
4
6
8
10
12
14
16
18
20
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Tem
peratu
re °
C
High status rivers
Impacted rivers
0
1
2
3
4
5
6
7
8
9
10
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
pH
(p
H u
nit
s)
High status rivers
Impacted rivers
226
Conductivity and alkalinity levels varied considerably during the study
programme in two of the high status sites, namely in the Brusna River and
in particular in the Dunneill River (Fig. 5.5 and 5.6 respectively). These
variations were not observed to the same extent in the other three high status
sites.
Conductivity levels reached a high of 588 µS/cm on the 26th
March 2002 in
the Dunneill River, which is an unexpectedly high value but it followed a
long spell of dry weather – only 2.3mm rainfall fell during the previous
week at the Straide rainfall station – effectively this was the first extended
dry period in 2002. The high conductivity value was not accompanied by,
for example, high B.O.D., ammonia or nutrient concentrations that might
indicate a pollution incident. Minimum levels of 73 µS/cm were recorded in
both January and July 2002 in the Dunneill and mean values for
conductivity were 260.5 µS/cm throughout the study (Appendix 5.1).
Fig. 5.5 Variation in mean conductivity levels in samples of stream water
collected on a monthly basis from the high status and impacted rivers during
the study.
0
100
200
300
400
500
600
700
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Co
nd
uct
ivit
y (
µS
/cm
)
High status rivers
Impacted rivers
227
Fig. 5.6 Variation in mean alkalinity levels in samples of stream water
collected on a monthly basis from the high status and impacted rivers during
the study.
Mean values for alkalinity in the Dunneill River (high status) were 104.4
mg/l CaCO3 and maximum concentrations of 298 mg/l CaCO3 were
recorded in August 2001. Concentrations ranged between 30-91 mg/l
CaCO3 from March to July 2002 and rose dramatically to 162 mg/l CaCO3
in August 2002 where they remained at a similar concentration until the end
of the study in October 2002. Mean alkalinity concentrations in the Brusna
River (high status) were 230.7 mg/l CaCO3 (Appendix 5.1). One of the
highest alkalinity values was recorded in this river in September 2001 (292
mg/l CaCO3) and it dropped to 136 mg/l CaCO3 in October 2001. A strong
relationship was apparent between alkalinity in the Dunneill and preceding
rainfall when graphed against 7-day cumulative rainfall for the day of
sampling (Fig. 5.7). This was possibly due to groundwater influence during
low flow periods.
0
50
100
150
200
250
300
350
400
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Alk
ali
nit
y m
g/l
(Ca
CO
3)
High status rivers
Impacted rivers
228
Alkalinity v Preceding Rainfall
y = -18.658Ln(x) + 99.703
R2 = 0.6369
-10
0
10
20
30
40
50
60
70
0 50 100 150 200 250 300 350
Alkalinity (mg/l CaCO3)
Cu
mu
lati
ve
7-d
ay
Ra
infa
ll
(mm
)
Fig. 5.7 Relationship between alkalinity and preceding rainfall in the
Dunneill River against 7-day cumulative rainfall for the day of sampling.
Alkalinity concentrations and conductivity levels fluctuated in the
Owengarve River (high status) also but not to the same extent as in the other
two rivers. Alkalinity concentrations rose from 130 mg/l CaCO3 in August
2001 to 252 mg/l CaCO3 in September 2001.
Conductivity and alkalinity also varied in the Robe River (impacted site) but
to a lesser extent (Fig. 5.5 and 5.6). These parameters did not vary
considerably in the remaining four impacted sites or in the other two high
status sites (Callow Loughs Stream and the Castlebar River).
The high alkalinity in the Robe and the Mullaghanoe Rivers (impacted river
-mean values of 308.3 and 159.2 mg/l CaCO3 respectively) and mean
conductivity levels of 594.0 µS/cm (Robe River) and 395.8 µS/cm
(Mullaghanoe River) reflected the calcareous nature of the catchment
geology especially in the vicinity just upstream of the site sampled and
resultant hard waters of these rivers. The Cartron, the Mad River and Lough
na Corralea Stream (impacted rivers) were all soft-water rivers having
catchments with predominantly siliceous bedrock geology with mean
alkalinity levels ranging from 8.4 to 18.2 mg/l CaCO3. In four out of the
five high status rivers the mean alkalinity values ranged from 104.4 to 230.7
mg/l CaCO3. The Castlebar River (high status) had the lowest alkalinity
229
level (mean value of 42.9 mg/l CaCO3) among the high status rivers
reflecting its soft-water nature.
In general, the variation in conductivity levels and alkalinity concentrations
reflects the catchment geology and variation in the discharges among sites.
5.4.1.3 Dissolved Oxygen (% saturation) and BOD (mg/l O2)
Mean values ranged from 101.1 to 107.2 % in the high status sites. On
occasion however, the DO was elevated (>110%) in some of the high status
rivers, particularly in the Owengarve and the Brusna Rivers (Appendix 5.1
and Fig 5.8). Mean values in the impacted sites ranged from 97.4 to 110.7%
during the study period. The Robe (impacted river) displayed the highest
DO level (158% saturation) among all ten rivers on 25th
September 2001
(Appendix 5.2). This is taken to indicate a greater degree of eutrophication
in the impacted river sites leading to higher levels of oxygen supersaturation
during daylight hours as a result of more intense photosynthesis.
Fig. 5.8 Variation in mean dissolved oxygen levels in samples of stream
water collected on a monthly basis from the high status and impacted rivers
during the study.
BOD levels were slightly higher in the impacted sites compared to the high
status sites (Fig. 5.9). Concentrations fluctuated considerably in Lough na
Corralea Stream (impacted) ranging from 0.4 to 3.5 mg/l O2. Levels also
0.0
20.0
40.0
60.0
80.0
100.0
120.0
140.0
160.0
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Dis
solv
ed
oxygen
(%
satu
rati
on
)
High status rivers
Impacted rivers
230
fluctuated in the Mullaghanoe River (impacted) ranging from 0.3 to 2.5 mg/l
O2 during the study. Mean values varied from 0.5 to 0.8 mg/l O2 in the high
status rivers and from 0.8 to 1.1 mg/l O2 in the impacted rivers.
Fig. 5.9 Variation in mean BOD concentrations in samples of stream water
collected on a monthly basis from the high status and impacted rivers during
the study.
5.4.1.4 Colour (Hazen units) and Chloride (mg/l Cl)
Colour appeared to vary more in the high status sites than in the impacted
sites (Fig 5.10). Mean values in the high status rivers ranged from 53.3 in
Callow Loughs Stream to 109.5 in the Castlebar River (Fig 5.10; Appendix
5.1). Mean values in the impacted sites ranged from 43.9 in the
Mullaghanoe River to 145.8 in the Cartron River (Appendix 5.2).
Fig. 5.10 Variation in mean colour measurements in samples of stream
water collected on a monthly basis from the high status and impacted rivers
during the study.
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Bio
chem
ica
l O
xy
gen
Dem
an
d
(mg
/l O
2)
High status rivers
Impacted rivers
0
50
100
150
200
250
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Colo
ur (
Hazen
un
its)
High status rivers
Impacted rivers
231
Chloride levels showed a narrow range in values in most rivers (mean
values from 10.1 to 24.5 mg/l Cl (Fig. 5.11; Appendix 5.1 and 5.2). The
Cartron River (mean value of 24.5 mg/l Cl) is located near the sea and that
may explain the broad range in values recorded here. Atlantic gales are
known to cause elevated chloride levels in inland rivers.
Fig. 5.11 Variation in mean chloride concentrations in samples of stream
water collected on a monthly basis from the high status and impacted rivers
during the study.
0
5
10
15
20
25
30
35
40
45
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Ch
lori
de
(mg
/l C
l)
High status rivers
Impacted rivers
232
5.4.1.5 Ammonia (mg/l N)
Mean concentrations of ammonia in the high status and impacted rivers are
shown in Fig 5.12. Ammonia concentrations maintained a level of 0.01mg/l
N in the Dunneill River (high status) throughout the study with the
exception of an elevated measurement of 0.11mg/l N that was recorded in
September 2001 (Appendix 5.1). Elevated levels (0.04 and 0.07mg/l N)
were detected in the Owengarve River (high status) on occasion indicating
perhaps that these rivers are not entirely pristine in nature. The
concentration of ammonia in the other three high status sites remained low
throughout the study (Appendix 5.1).
Fig. 5.12 Variation in mean ammonia concentrations in samples of stream
water collected on a monthly basis from the high status and impacted rivers
during the study.
Elevated levels were detected in some of the impacted rivers also (Fig.
5.12). The highest concentration (0.21 mg/l N) was recorded in the
Mullaghanoe River (impacted river) in August 2002 and displayed a mean
of 0.08 mg/l N throughout the study period (Appendix 5.2). High
concentrations were also detected in Lough na Corralea Stream (impacted
river) with values ranging from 0.02 to 0.1 mg/l N (Appendix 5.2).
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Am
mon
ia (
mg/l
N)
High status rivers
Impacted rivers
233
5.4.1.6 TON (mg/l N)
TON concentrations were elevated on occasion in some of the high status
rivers and fluctuated from 0.3 to 1.4 mg/l N in the Owengarve River and
from 0.2 to 0.95 mg/l N in the Brusna Rivers (Fig. 5.13). The median, upper
quartile and 90 percentile values for TON for 99 rivers in the West of
Ireland (1996-2001) were 0.687, 1.286 and 2.024 mg N/l respectively
(Appendix 5.4). Concentrations in the other three high status rivers ranged
from 0.1 to 0.5 mg/l N.
Fig. 5.13 Variations in mean TON concentrations in samples of stream
water collected on a monthly basis from the high status and impacted rivers
during the study.
The highest TON levels in the impacted rivers were recorded in the Robe
(1.6 mg/l N) and the Mullaghanoe Rivers (1.4 mg/l N) (Appendix 5.2).
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Month
Tota
l O
xid
ised
Nit
rogen
(m
g/l
N)
High status rivers
Impacted rivers
234
5.4.1.7 Unfiltered MRP (mg/l P)
Fluctuations in the MRP concentrations in both the high status and impacted
rivers during the course of the study period are shown in Fig 5.14. Mean
values ranged from 0.032 mg/l P to 0.042 mg/l P in the high status rivers
throughout the study (Appendix 5.1). In comparison with 99 rivers sampled
over the period 1996 to 2001 in the West of Ireland by the EPA, Castlebar,
the mean, median, 75 and 90th
percentile MRP concentrations were 0.034,
0.018, 0.032 and 0.050 mg P/l (Appendix 5.4). Elevated levels of MRP
were recorded in the Castlebar River (0.098 mg/l P), the Brusna River
(0.105 mg/l P), the Dunneill River (0.112 mg/l P) and Callow Loughs
Stream (0.082 mg/l P) during September 2001 (Fig. 5.1). Interestingly,
these high levels appeared to coincide with increased conductivity and
alkalinity values for the same period (Section 5.4.1.2). These appear to
coincide with a low flow period in September 2001.
Fig. 5.14 Variations in mean unfiltered MRP concentrations in samples of
stream water collected on a monthly basis from the high status and impacted
rivers during the study.
Mean MRP concentrations in the impacted rivers ranged from 0.010 to
0.053 mg/l P (Appendix 5.2). The Mullaghanoe River displayed the highest
recorded value of 0.098 mg/l P in August 2001 (Appendix 5.2).
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Un
filt
ered
MR
P (
mg/l
P)
High status rivers Mean
Impacted rivers Mean
235
5.4.1.8 Statistical comparisons of physico-chemical results between the
high status v impacted sites
The physico-chemical data were transformed where appropriate and a
repeated measures analysis (general linear model, SPSS) was carried out to
assess if any significant differences in the parameters existed between the
high status and impacted sites. Results are outlined in Table 5.6.
236
Table 5.6 Comparisons of the physico-chemical parameters between the high status and impacted sites using the general linear model in
repeated measures (SPSS).
Parameter
Mean concentration compared
Source of variation (between groups)
High status vs Impacted rivers
Source of variation
(within-subjects effects)
Month Month*status
pH (pH units) Reference < Impacted F1,3 = 3.239
P = 0.170
F8,24 = 1.343
P = 0.271
F8,24 = 0.403
P = 0.908
Temperature (°C)
F1,3 = 0.903
P = 0.412
F7,21 = 15.126
P = 0.001***
F 7,21 = 0.409
P = 0.886
Colour (Hazen units)
F1,3 = 0.767
P = 0.446
F10,30 = 5.467
P = 0.001***
F10,30 = 1.383
P = 0.235
Chloride (mg/l Cl)
F1,2 = 4.740
P = 0.161
F9,18 = 2.337
P = 0.06
F9,18 = 0.397
P = 0.921
Conductivity (µS/cm2)
Reference < Impacted F1,3 = 4.104
P = 0.136
F9,27 = 0.764
P = 0.650
F9,27 = 0.721
P = 0.685
Alkalinity (mg CaCO3)
Reference < Impacted F1,2 = 0.710
P = 0.488
F9,18 = 1.984
P = 0.103
F9,18 = 0.619
P = 0.766
Dissolved Oxygen (% O2 saturation)
F1,3 = 0.492
P = 0.724
F7,21 = 1.266
P = 0.314
F7,21 = 1.103
P = 0.397
Biochemical Oxygen Demand (mg/l O2)
F1,2 = 2.435
P = 0.259
F7,14 = 2.312
P = 0.086
F7,14 = 0.496
P = 0.822
Total Oxidised Nitrogen (mg/l N)
F1,2 = 0.136
P = 0.747
F9,18 = 1.827
P = 0.132
F9,18 = 1.107
P = 0.406
Unfiltered MRP (mg/l P)
Reference > Impacted F1,2 = 0.06
P = 0.83
F9,18 = 0.902
P = 0.544
F9,18 = 0.444
P = 0.893
Ammonia (mg/l N)
Reference < Impacted F1,3 = 1.534
P = 0.001***
F7,21 = 28.337
P = 0.001***
F7,21 = 32.548
P = 0.001***
p-values:
* p<0.05; ** p<0.01; *** p<0.001
237
Significant differences in ammonia concentrations (p=0.001), an indicator
of organic pollution, were detected between the high status and impacted
sites. There were no other significant differences in the physico-chemical
parameters between the high status and impacted rivers. Somewhat
unexpectedly, the mean MRP values in the ‘high’ status rivers were higher
than those recorded in the ‘impacted’ sites which was mainly due to low
N:P ratios (see Table 5.7 below). It appears that many of the clean rivers in
the study were N limited for considerable periods.
The MRP concentrations were elevated on occasion in some of the high
status rivers, particularly in the Castlebar and the Brusna Rivers, causing a
high variance during the study period (Fig. 5.14). In September 2001, the
MRP concentration in the Brusna River was 0.105 mg/l P and this
contributed to the overall high variance in this river. On removing the MRP
value for this month, the average concentration across the sampling period
was 0.036 mg/l P. During the same month, the MRP level taken in the
Castlebar River was also high, measuring 0.098 mg/l P, again, increasing
the variance across the study. Again, when this value was eliminated, the
average MRP concentration across the study was 0.032 mg/l P. Due to the
unexpectedly high MRP values in the high status sites and to the suspicion
that they were not actively P-limited, it was decided to take a closer look
and investigate the variation in the N:P ratios in each river over the entire
sampling period (Table 5.7).
5.4.1.9 Investigation in N:P ratios between high status v impacted sites
The N:P ratios fluctuated on a monthly basis between the study sites (Table
5.7). The variation between the high status and impacted sites is shown in
Fig. 5.15.
238
Table 5.7 Monthly variation in the N:P Ratios recorded in the high status and impacted rivers during the study period.
N:P ratios <15 points to an N limited system (values in bold).
Reference rivers
Impacted rivers
Owengarve Castlebar Brusna Dunneill Callow Loughs
Stream
Cartron Lough na Corralea
Stream
Robe Mullaghanoe Mad
May-01 75 32
Jun-01 7 36 34
Jul-01 27 13 19 14 19 33 35 14
Aug-01 6 13 24 28 42 11
Sep-01 62 4 7 2 5 38 70 38 33 5
Oct-01 30 14 35 20 15 57 53 41
Nov-01 54 80
Dec-01 66 27 54 32 32 88
Jan-02 32 19 11 21
Mar-02 71 17 57 20 29 35 55 89 71
Apr-02 61 9 49 14 26 57 49
May-02 20 9 16 13 26 20 63
Jun-02 24 9 22 10 17 16 38 40 33 34
Jul-02 25 7 17 87 14 17 49 40 13
Aug-02 22 4 8 9 15 13 31 5
Sep-02 47 13 35 15 23 8 44 31 66 8
Oct-02 26 8 23 12 15 10 22 30 41 11
239
Fig. 5.15 Variation in the N:P ratios between the high status and impacted
rivers during the sampling programme.
Fig. 5.16 Box and whisker plot showing the variation in the NP ratios in the
ten rivers during the sampling period (p =0.001).
N
P R
ati
os
-20
0
20
40
60
80
100
Reference rivers Impacted rivers
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
0
10
20
30
40
50
60
70
80
90
100
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
N:P
rati
os
High status rivers
Impacted rivers
240
The N:P ratios did vary considerably between the high status and impacted
rivers (Fig. 5.15). The lowest N:P ratios were found in the Castlebar River
and Callow Loughs Stream, both high status rivers. With the exception of
one sampling occasion (August 2001, Table 5.7), the Owengarve River was
P-limited for the entire sampling period. The same was true for the Brusna
River, which was P-limited throughout aside from being N-limited in
September 2001. The Castlebar River was practically N-limited right
throughout the study apart from December 1001 and March 2002 (Table
5.7, see also Chapter 2 split stream study). The Dunneill River fluctuated
considerably on a month to month basis between N and P-limitation (Table
5.7). Callow Loughs Stream was N-limited from July to September 2001
inclusive and switched to a P-limited system from December 2001 to June
2002. As with the previous year, the river became N-limited in July and
August 2001. In September and October 2002 the system changed back to
being P-limited.
In terms of the impacted sites, the Mad River was N-limited on seven of the
eleven sampling occasions and was the dominant impacted river to display
N-limitation. With the exception of three sampling occasions in the Cartron
River and one in Lough na Corralea Stream, four out of five of the impacted
rivers remained P-limited for the duration of the study (Table 5.7).
In general, the impacted rivers were found to be more P-limited during the
study in comparison to the high status rivers (Fig. 5.16; Table 5.7). The box
and whisker plot in Fig. 5.16 compares the N:P ratios between the high
status and impacted rivers throughout the study showing an overall
significant difference (p<0.001) when a one-way ANOVA was applied to
the data. In general, the N:P ratios in the high status rivers, were slightly
lower during the summer of 2001 and 2002, compared to the other seasons
(Table 5.7). Overall, in the impacted rivers, the N:P ratios were similar in
the summer and autumn months. In some of the impacted rivers, however,
the ratios seemed higher during winter 2001/2002 and spring 2002. A
similar pattern was also observed in the high status sites during these
seasons. The results here and also those in Chapter 2 demonstrate that some
higher status rivers may be nitrogen limited particularly during the late
241
summer months rather than the more generally found P-limitation. An
analysis of the EPA database of the N:P ratios for 99 rivers in the West of
Ireland found that approximately 4% of samples analysed are N-limited and
with low MRP concentrations (<0.05 mg/l P; Appendix 5.4). There is also a
definite tendency for nitrogen limitation to occur during the summer months
– a Chi-Square analysis of observed versus expected frequency of
occurrence of N limitation at low MRP concentrations suggested a strong
bias towards the summer months (Fig. 5.17). An initial analysis of the
catchment characteristics of sites exhibiting low N:P ratios with low MRP
concentrations has not revealed any obvious controlling factor such as
geology or particular forms of land use.
Chi -Squared Comparison Occurrence of N-Limitation at Low MRP
Concentrations
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
1 2 3 4 5 6 7 8 9 10 11 12
Month
Pe
rce
nta
ge
of
Sa
mp
les
All Samples
N-limited Samples
Fig. 5.17 Occurrence of N-limitation at low MRP concentrations in 99
rivers in the West of Ireland (1996-2001). N-limitation is more likely to
occur between July and October (X2 = 54.9, p<0.001).
242
5.4.2 Night time dissolved oxygen (DO) measurements
It was important to establish whether minimum dissolved oxygen saturation
during the hours of darkness could be an important factor controlling the
occurrence of Ecdyonurus. Table 5.8 outlines the results for the night-time
dissolved oxygen saturation measurements taken in August 2003 during low
flow, warm conditions.
Table 5.8 Percentage Saturation of dissolved oxygen and temperature
readings obtained during the night-time hours.
Sampling date Time % DO
saturation
DO
mg/l O2
Temperature
°C
High status Rivers
Owengarve 7th
August 2003 02.40hrs 80 7.3 19.6
Castlebar 7th
August 2003 01.40hrs 92 8.7 17.5
Brusna 7th
August 2003 04.00hrs 88 8.2 18.0
Dunneill 7th
August 2003 03.45hrs 90 8.4 18.6
Callow Loughs Stream 7th
August 2003 05.00hrs 91 8.5 16.7
Impacted Rivers
Cartron 14th
August 2003 04.15hrs 92 9.4 13.1
Lough na Corralea Stream 7th
August 2003 00.05hrs 85 7.5 20.2
Robe 7th
August 2003 00.40hrs 85 7.9 21.0
Mullaghanoe 7th
August 2003 02.10hrs 65 6.1 16.8
Mad 7th
August 2003 03.15hrs 97 8.9 19.0
The National Salmonid Waters Regulations incorporating the requirements
(78/659/EEC) of the Freshwater Fish Directive require that 50% of samples
should be greater the 9 mg/l O2 and that all samples should be greater than 6
mg/l O2. Thus, even though only one measurement was taken during one
night-time visit, the concentration in the Mullaghanoe River (impacted
river; 6.1 mg/l O2) is close to breaching these requirements. The DO
concentration in the Owengarve River (high status river; 7.3 mg/l O2), in
Lough na Corralea Stream (Impacted river; 7.5 mg/l O2) and in the Robe
River (Impacted river; 7.9 mg/l O2) was also less than the 9 mg/l O2. More
frequent sampling would be required to demonstrate the pattern of night
time oxygen depletion and particularly at the lowest flow conditions during
the growing seasons. Unfortunately this was beyond the scope of the
current study.
243
5.4.3 Results of the macroinvertebrate analysis
The diversity indices and metrics were examined on a river-by-river basis
across the 16-month sampling period (Section 5.4.3.1 to 5.4.3.7). The
graphs in Figs. 5.18 to Fig 5.24 show the variation in these indices/metrics
between the high status and impacted rivers. Table 5.9 outlines the mean
values for the various indices/metrics applied to average density/m2
macroinvertebrate data in each of the ten rivers. The variation in the
diversity indices and biotic metrics was examined on a river-by-river basis
across five seasons and are presented in Appendices 5.5 to 5.11 (Section
5.4.3.8). The diversity indices and metrics were compared (Kruskal-Wallis
ANOVA) between the high status and impacted sites (Appendix 5.12; Table
5.10).
Table 5.9 Mean values for the various metrics/indices calculated in the high
status and impacted rivers during the sampling period.
High status rivers
Shannon-
Wiener
index (H’)
Margalef’s
index (d)
Pielou’s
evenness
(J’)
Simpsons
index
Total
number of
taxa (S)
Total
number of
individuals
(N)
%
EPT
Owengarve 2.9 5.1 0.8 0.9 38 1625 0.43
Castlebar 2.1 3.3 0.7 0.8 23 1004 0.49
Brusna 1.81 4.2 0.5 0.7 34 2639 0.44
Dunneill 2.4 4.1 0.7 0.9 30 1375 0.36
Callow Loughs Stream 2.4 4.7 0.7 0.8 36 1703 0.49
Impacted rivers
Cartron 2.1 2.7 0.7 0.7 20 1251 0.44
Lough na Corralea stream 2.5 3.6 0.8 0.9 26 1269 0.36
Robe 1.7 2.0 0.6 0.7 17 2823 0.27
Mullaghanoe 2.1 3.3 0.6 0.8 30 7913 0.35
Mad 1.9 2.6 0.7 0.8 17 454 0.23
The AQEM software was also applied to average density/m2
data to
calculate a wider range of biotic indices to further compare the high status
and the impacted sites (Section 5.4.3.9; Appendix 5.13).
244
Table 5.10 Results for the Kruskal-Wallis ANOVA applied to the diversity
indices and metrics. The variation between the high status and impacted
rivers were examined.
Index/Metric Reference x impacted rivers
p-value Chi-square
Shannon-Wiener diversity (H’)
P=0.239 ns 1.386
Margalef’s index (d)
p<0.001*** 34.662
Pielou’s evenness (J’)
P=1 ns 0
Simpsons diversity (λ)
p=0.239 ns 1.386
Total number of taxa (S)
p<0.001*** 12.478
Taxon abundance (N)
P=0.239 1.386
Percentage EPT
p<0.001*** 19.689
p-values: *p<0.05; **p<0.01; ***p<0.001; ns – not significant
5.4.3.1 Shannon-Wiener Index (H’)
Mean values for Shannon-Wiener index were slightly higher in the high
status rivers compared to the impacted rivers (Fig. 5.18; Table 5.9). No
significant difference (Table 5.10; p=0.239) between the high status and
impacted sites was observed when compared using this index.
Fig. 5.18 Variation in the Shannon-Wiener diversity index (mean values) in
the high status and impacted sites during the study.
0
0.5
1
1.5
2
2.5
3
3.5
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Nov
-02
Date
Sh
an
non
-Wie
rn
er I
nd
ex (
H')
High status rivers
Impacted rivers
245
5.4.3.2 Margalef’s Index (d)
Mean values for Margalef’s index (d) were generally higher in the high
status than in the impacted rivers (Fig. 5.19) and ranged from between 2 to 5
(Table 5.9). A highly significant difference (Table 5.10; p<0.001) was
found between the two groups of rivers using Margalef’s index.
Fig. 5.19 Variation in Margalef’s index (mean values) in the high status and
impacted sites during the study.
0
1
2
3
4
5
6
7
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Nov
-02
Date
Margale
f's
Ind
ex (
d)
High status rivers
Impacted rivers
246
5.4.3.3 Pielou’s evenness (J’)
The variation in Pielou’s evenness (J’) is shown in Fig. 5.20 with mean
values ranging from 0.5 to 0.8 (Table 5.9). There was no significant
difference (Table 5.10; p=1) between the high status and impacted sites
using Pielou’s evenness.
Fig. 5.20 Variation in Pielou’s evenness index (mean values) in the high
status and impacted sites during the study.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Nov
-02
Date
Pie
lou
's e
ven
ness
(J')
High status rivers
Impacted rivers
247
5.4.3.4 Simpsons diversity
Mean values for Simpsons index ranged from 0.7 to 0.9 (Fig. 5.21; Table
5.9). When comparing the two groups of rivers using this index, no
significant difference was observed (Table 5.10; p=0.239)
Fig. 5.21 Variation in Simpsons index (mean values) in the high status and
impacted sites during the study.
0
0.2
0.4
0.6
0.8
1
1.2
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Nov
-02
Date
Sim
pso
ns
div
ersi
ty
High status rivers
Impacted rivers
248
5.4.3.5 Total number of taxa (S)
Mean values for the total number of taxa were generally higher in the high
status than in the impacted sites (Fig. 5.22; Table 5.9). The high status and
impacted sites were significantly different (Table 5.10; p<0.001) when
compared using total number of taxa. The lowest mean value (17) was
found in the Mad River and the Robe Rivers (impacted river) while the
highest mean value (38) originated in the Owengarve River (high status
river). It is worth noting that the Mullaghanoe (impacted by Charlestown
sewage works) had a greater number of taxa than the Castlebar River site,
which is recognised as perhaps one of the highest status sites in the study.
Thus, reliance on number of taxa as a simple index of ecosystem health is
not advisable in isolation. The number of invertebrate taxa present in rivers
can increase in the early stages of eutrophication or organic pollution as, for
example, the number of gastropod species increases in response to increased
primary production. The addition of more tolerant grazers while at the same
time some of the more sensitive taxa are still able to survive will boost the
overall number of taxa as a response to eutrophication in this case.
Fig. 5.22 Variation in the total number of taxa (mean values) in the high
status and impacted sites during the study.
0
1
2
3
4
5
6
7
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Nov
-02
Date
Tota
l n
um
ber o
f ta
xa (
S)
High status rivers
Impacted rivers
249
5.4.3.6 Taxon abundance (N)
The mean taxon abundance (N) (Fig. 5.23; Table 5.9) was quite similar in
the Owengarve River (1624.9), the Castlebar River (1004.4) and the
Dunneill River (1374.9). No significant differences (Table 5.10; p=0.239)
were observed between the high status and impacted sites using this index.
The Brusna River and Callow Loughs Stream contained slightly higher
mean numbers (2639.3 and 1702.7 respectively). The highest abundance of
invertebrates in all ten rivers was found in the Robe River (2823.2) and the
Mullaghanoe River (7913.3) (both impacted sites).
Fig. 5.23 Variation in the total number of individuals N (Abundance) in the
high status and impacted sites during the study (mean values).
0
2000
4000
6000
8000
10000
12000
14000
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Date
Taxon
ab
un
dan
ce (
N)
High status rivers
Impacted rivers
250
5.4.3.7 Percentage EPT
The highest mean percentage EPT value of 0.5 was found in the Castlebar
River (high status river) while the Mad River displayed the lowest mean
value of 0.2 (Fig. 5.24; Table 5.9). Significant differences were observed
(p<0.001) between the high status and impacted sites when compares using
percentage EPT. Generally speaking, with the exception of the Cartron
River (impacted site), the mean percentage EPT was higher in the high
status rivers compared to the impacted rivers. The Cartron River had a mean
value of 0.4, which was quite high compared to the other impacted sites due
to the high numbers of Baetis spp., Leuctra spp., and other plecopteran and
trichopteran species in this river.
Fig. 5.24 Variation in percentage EPT (mean) in the high status and
impacted sites during the study.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
May
-01
Jul-0
1
Sep-0
1
Nov
-01
Jan-
02
Mar
-02
May
-02
Jul-0
2
Sep-0
2
Nov
-02
Date
% E
PT
High status rivers
Impacted rivers
251
5.4.3.8 Seasonal variations
The variation in the diversity indices and biotic metrics examined on a river-
by-river basis across five seasons results are presented in Appendix 5.5 to
5.11). The data for November 2001, December 2001 and January 2002 was
omitted as only kick samples were taken during these months which would
have been inconsistent with Surber data collected for all other seasons.
These plots give an indication of the variation in the range of indices and
matrices among seasons. Not all samples were sorted in the impacted sites
thereby creating some gaps in the seasonal data.
On a river-by-river basis, there were no substantial differences in the
diversity indices between seasons in the high status sites (Appendix 5.5 to
5.11). Species richness (d) dropped slightly in the Dunneill River and
Callow Loughs Stream in autumn 2001 (Appendix 5.5). The opposite was
observed in the Owengarve, Castlebar and Brusna Rivers (high status)
where species richness increased slightly during the same period. Different
insect emergent patterns among the various species may have contributed to
these variations. It would normally be expected that plecopteran species, in
particular, would reappear in the benthic faunal macroinvertebrate
community from early September onwards but typically many species will
be absent during the summer months. The emergence pattern of
Rhithrogena also typically will mean that few if any nymphs of this genus
will be present from mid July to late August.
The total number of taxa/m2
and the total number of individuals/m2
(Appendix 5.6 and 5.7 respectively) increased slightly in spring 2002 in
Callow Loughs Stream. An increase in the total number of taxa/m2 in the
Castlebar, Brusna and Dunneill Rivers was also observed during this season
but not as pronounced as in Callow Loughs Stream (Appendix 5.5). The
Owengarve on the other hand appeared to maintain a consistently high
number of taxa during the studies with little change between seasons
highlighting the homogeneity of this system.
252
The highest percentage EPT in the high status rivers was found in autumn
2002 in the Castlebar River (Appendix 5.8). A similar proportion was
found in autumn 2001 in Callow Loughs Stream. The percentage EPT did
not fluctuate dramatically between seasons and across sites. Percentage
EPT also demonstrated strong differences between impacted and high status
sites suggesting that, in combination with its consistency across seasons, it
may be quite a useful indicator of ecosystem health in Irish rivers.
The Mullaghanoe River was the most diverse of all the impacted sites
(Appendix 5.9 and 5.10). Species richness (Appendix 5.5) was highest in
the Mullaghanoe River, particularly during summer 2001, 2002 and autumn
2002. Abundance (N) was also highest in summer 2001, summer 2002 and
to a lesser extent in autumn 2002 (Appendix 5.7). Percentage EPT was
generally low but consistent across all five seasons in this impacted river.
Generally speaking, there was an increase in species richness during the
summer 2002 in three of the five impacted rivers (Appendix 5.5). Tolerant
species generally increase in impacted rivers therefore pollution impacts
coupled with changes in species life cycles may be driving these changes.
Taking gaps in the dataset into consideration, with the exception of the Mad
River, the diversity indices examined in the other three impacted rivers were
generally consistent across seasons. As with the Mullaghanoe River, the
total number of taxa in the Cartron River was highest in spring 2002
(Appendix 5.6). The total number of individuals/m2 was quite consistent
across the impacted sites (with the exception of the Mullaghanoe and the
Robe rivers). The numbers increased in spring 2002 in the Robe River
while they dropped in the Mullaghanoe River during the same period
(Appendix 5.7).
There did not appear to be any variation in Shannon-Wiener index
(Appendix 5.9), Simpsons diversity (Appendix 5.10) or in Pielou’s evenness
(Appendix 5.11) in the high status or impacted rivers across the seasons
studied.
253
The diversity indices and biotic metrics were not as consistent across the
five seasons in the impacted sites as they were in the high status sites. Gaps
in the data set obviously contributed to a lack of information for particular
seasons but it was still possible to gather knowledge of seasonal patterns
within the community structure of these impacted rivers.
5.4.3.9 Calculation of AQEM metrics
The macroinvertebrate average numbers/m2 data were analysed using the
AQEM software (AQEM V2.3.4a, 2004) producing a series of
metrics/indices.
A one-way ANOVA was applied to a selection of these metrics/indices,
showing significant differences in all selected indices/metrics between the
high status and the impacted sites (Appendix 5.13; Table 5.11). A summary
of the p-values and the means recorded from comparisons (ANOVA) made
between the high status and impacted sites for the indices/metrics
(Appendix 5.13), microhabitat preferences (Appendix 5.14) and feeding
types (Appendix 5.15) derived from the AQEM software is presented in
Table 5.11.
These metrics focus mainly on organic pollution and eutrophication and
based on the occurrence of the macroinvertebrate communities in the high
status and impacted sites, the status of these rivers were confirmed with
clear differentiation between the two groups.
The high status rivers were more species rich and contained more sensitive
indicator taxa of organic pollution and eutrophication. The impacted sites,
however, while only being moderately impacted (see BBI, IBE, ASPT in
Appendix 5.13) did contain a higher percentage of tolerant taxa in
comparison to the higher status sites.
As part of the AQEM assessment methodology a multimetric index is
produced which integrates multiple attributes of stream communities
(“metrics”) to describe and assess the health of a river system. Key areas of
interest included investigating the feeding measures and the microhabitat
254
preferences among the high status and impacted sites. Feeding measures
comprise functional feeding groups and provide information of the balance
of feeding strategies (food acquisition and morphology) in the benthic
assemblage. Microhabitat preferences are based on individual
macroinvertebrate tendencies to survive in a particular microhabitat type.
Table 5.11 Summary of the p-values and means recorded from comparisons
made between the high status and impacted sites for the indices/metrics,
microhabitat preferences and feeding types analysed from the AQEM
software.
Variable Status Mean p-values
High status sites vs
Impacted sites
Index/metric
Czech Saprobic Indec
High status sites
Impacted sites
1.3
1.5
p<0.001***
Dutch Saprobic Index High status sites
Impacted sites
0.2
0.1
p=0.009**
German Saprobic Index High status sites
Impacted sites
1.6
1.7
p<0.001***
Danish Stream Fauna Index (DSFI) High status sites
Impacted sites
6.7
5.8
p<0.001***
Indice Biotico Esteco (IBE) High status sites
Impacted sites
9.9
8.6
p<0.001***
Belgian Biotic Index (BBI) High status sites
Impacted sites
8.8
7.4
p<0.001***
Average Score per Taxon (ASTP) High status sites
Impacted sites
5.9
5.6
p<0.001***
Microhabitat preference
Lithal microhabitat type High status sites
Impacted sites
48.4
33.9
p<0.001***
Phytal microhabitat type High status sites
Impacted sites
26.4
21.2
p<0.001***
Pelal microhabitat type High status sites
Impacted sites
6.1
13.2
p<0.001***
Akal microhabitat type High status sites
Impacted sites
7.3
10.9
p<0.001***
Feeding type
Grazers and scrapers High status sites
Impacted sites
58.2
31.9
p<0.001***
Miners High status sites
Impacted sites
0.6
1.7
p<0.001***
Gatherers/collectors High status sites
Impacted sites
22.6
31.6
p<0.001***
Active filter feeders High status sites
Impacted sites
1.5
3.5
p<0.001***
Passive filter feeders High status sites
Impacted sites
4.0
11.7
p<0.001***
Predators High status sites
Impacted sites
4.5
7.1
p=0.002**
Parasite High status sites
Impacted sites
0.7
1.7
p<0.001***
p-values: * p<0.05; ** p<0.01; *** p<0.001.
255
5.4.3.10 Microhabitat preferences
Descriptions of the microhabitat types derived in the AQEM project are
outlined in Tables 5.12.
Table 5.12 Microhabitat types.
Microhabitat Type Description of microhabitat
Pel Pelal: mud; grain size < 0.063mm (%)
Arg Argyllal: silt, loam, clay; grain size < 0.063mm (%)
Psa Psammal: sand; grain size 0.063-2mm (%)
Aka Akal: fine to medium-sized gravel; grain size 0.2-2cm (%)
Lit Lithal: coarse gravel, stones, boulders; grain size > 2cm (%)
Phy Phytal: algae, mosses and macrophytes including living
parts of terrestrial plant (%)
POM Particulate organic matter, such as woody debris, CPOM
FPOM (%)
Oth Other habitats (%)
Significant differences in the Pelal microhabitat type (p<0.001), Akal
microhabitat type (p<0.001), Lithal microhabitat type (p<0.001) and Phytal
microhabitat type (p<0.001) were observed between the high status and the
impacted sites were observed (Table 5.12; Appendix 5.14).
A higher percentage of the invertebrate community in the impacted sites
showed a preference for the Pelal (mud-grain size <0.063mm) and the Akal
(fine to medium-sized gravel; grain size 0.2-2cm) microhabitat type
compared to the high status sites.
In comparison, the high status rivers contained significantly higher
percentages of invertebrate fauna preferring coarse gravel, stones and
boulders (Lithal microhabitat; grain size > 2cm), algae, mosses and
macrophytes including living parts of terrestrial plants (Phytal microhabitat)
compared to the impacted sites.
The AQEM software produces scores (X out of 10 points) for each
individual taxon or species in describing the autecological information:
10/10 being the most favoured and 0/10 being the least favoured for the
256
particular autecological feature expressed. As mentioned above, the
microhabitat and feeding types were of particular interest in this study.
Ecdyonurus dispar and Ecdyonurus insignis scored 10/10 and Ecdyonurus
venosus and the immature unidentified Ecdyonurus species scored 8/10 in
preference for the Lithal microhabitat type. Results show that the sensitive
indicator species like Ecdyonurus score highly in the rivers that contain
coarse gravel, stones and boulders. In addition to this, the high status sites
contain significantly higher percentages of taxa with a preference for this
microhabitat type indicating that these high status sites have the capacity to
sustain the more sensitive indicator species. Interestingly, sensitive
indicator taxa like Ecdyonurus, Perla and some Trichopteran species
preferred habitats (10/10) with gravel, stones and boulders only and are
deemed to have a zero weighting or zero preference (0/10) for microhabitats
containing mud, medium-sized gravel and habitats with algae, moss or
macrophytes.
Taxa like Chironomidae species score 6/10, Dicranota spp. scored 4/10,
Oligochaetea species 3/10 while Tipula species and Tubificidae species
scored 5/10 with a preference for the Pelal microhabitat (mud loving taxa).
In addition to this, these taxa all score 0/10 in preference to the Lithal
microhabitat type. Simulium species scored 5/10 (highest score among all
taxon) with a preference for the Akal microhabitat type (fine to medium-
sized gravel). The results demonstrate that the impacted sites contained a
higher percentage of mud loving taxa and these with a preference for fine to
medium-sized gravel (0.2-2cm).
This is seen as an important result - especially when compared with the
results based on particle size alone (Section 5.4.4.4). The particle size
analysis of substratum did not show any significant differences between the
high status sites and the impacted sites. When it is considered that all of the
impacted sites were selected on the basis that they once were capable of
supporting ‘lithal’ species such as Ecdyonurus – and indeed Ecdyonurus has
been historically recorded at all sites; albeit absent or almost absent during
this survey at the impacted sites. The distinct faunal community differences
257
relating to microhabitat preferences, however, suggest that quite subtle
changes in the microhabitat of the impacted sites has taken place. The
faunal community as a whole has apparently shifted from a Lithal-
dominated one to a community that has a higher proportion of Pelal species
and other taxa that prefer a finer-particle substratum.
5.4.3.11 Feeding types
Significant differences between the high status and impacted sites were
observed when comparing the feeding types (Appendix 5.15; Table 5.11).
The percentage of grazers and scrapers were significantly higher (Table
5.11; p<0.001) in the high status compared to the impacted sites. All other
feeding types outlined in Table 5.11 were significantly higher in the
impacted sites (Appendix 5.15) giving an indication into the feeding
strategies among the invertebrate fauna within the two groups.
This further supports the idea that biotic influences may be quite
significantly influence as eutrophication impacts progress. As with the
microhabitat metrics statistically significant differences are apparent
between the high status sites and the impacted sites. It is more difficult to
understand the precise significance of having more grazers and scrapers in
the ‘cleaner’ environment but it may be related to the condition of the
substratum surface and interstices at a microhabitat level. A clean
oligotrophic environment may favour grazers and scrapers whereas a more
productive eutrophic system becomes ‘clogged’ knocking out the species
which require clean stone surfaces with a modicum of algal growth to feed
upon. If growth rates exceed a certain threshold and perhaps if algal species
types change these invertebrate species are placed at a disadvantage and
other types begin to replace them.
258
5.4.4 Results of sediment analysis
Substrate fraction dry weights (g) for each site is presented in Appendix
5.16. The mean percentage frequency distribution (by weight) from the
individual sites is presented together with individual sample data in
Appendix 5.17 (high status sites) and Appendix 5.18 (impacted sites). The
results for the high status and impacted sites outlined below refer to the
calculated mean values presented in Appendices 5.17 and 5.18.
The main aim of this section of the study was to determine whether
significant differences in substrate composition were apparent when
impacted sites were compared with high status sites and thus, to attempt to
understand the importance of sediment and siltation in controlling the
distribution of Ecdyonurus in particular.
5.4.4.1 Substrate fractions - High status rivers
The substrate fractions in the high status sites are presented in Figs. 5.25 to
5.29. The Dunneill River contained all particle sizes while samples from
the other four high status sites had most size fractions with the exception of
boulders (256-512mm) (Figs. 5.25 to 5.29). All rivers contained small
cobble (64-128mm) and large cobble (128-256mm). Small cobble was the
dominant particle size (by weight) in the Dunneill River, the Castlebar
River, Callow Loughs Stream and the Brusna River, ranging from 45.01%,
48.65%, 40.62% and 45.06% respectively (Appendix 5.17). The Brusna
River contained the highest percentage of large cobble among the five high
status sites, representing 34.12% of the total sample weight. The other four
river sites contained large cobble in smaller amounts ranging from 13.80%
to 18.42%. The dominant fraction in the Owengarve River was coarse
gravel accounting for 22.31% of the total fraction. Very coarse gravel (32-
64mm)and large and small cobble were also recorded in this site accounting
for 12.69%, 11.58% and 14.22% respectively of the total fraction weight.
While carrying out routine monthly sampling, visual observations noted that
this river regularly contained quite a lot of silt and sand. The silt or fine
sediments (particles <1mm diameter) are examined separately in both the
high status and impacted sites in section 5.4.4.4.
259
Fig. 5.25 Mean frequency particle size distribution (by % weight) from the
Owengarve River in March 2003. Error bars indicate maximum and
minimum values.
Fig 5.26 Mean frequency particle size distribution (by % weight) from the
Dunneill River in March 2003. Error bars indicate maximum and minimum
values.
Fig. 5.27 Mean frequency particle size distribution (by % weight) from the
Castlebar River in March 2003. Error bars indicate maximum and minimum
values.
0
10
20
30
40
50
60
70
80
90
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
12
8
25
6
Sieve size (mm)
Fre
qu
ency
(%
)
Sand GravelCobble
SiltBoulders
0
10
20
30
40
50
60
70
80
90
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
128
256
Sieve size (mm)
Fre
qu
ency
(%
)
Silt Sand Gravel Cobble
0
10
20
30
40
50
60
70
80
90
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
128
256
Sieve size (mm)
Fre
qu
ency
(%
)
Silt Sand Gravel Cobble
260
Fig. 5.28 Mean frequency particle size distribution (by % weight) from the
Callow Loughs Stream in March 2003. Error bars indicate maximum and
minimum values.
Fig. 5.29 Mean frequency particle size distribution (by % weight) from the Brusna River in
March 2003. Error bars indicate maximum and minimum values.
0
10
20
30
40
50
60
70
80
90
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
128
256
Sieve size (mm)F
req
uen
cy (
%)
Silt Sand Gravel Cobble
0102030405060708090
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
128
256
Sieve size (mm)
Fre
qu
ency
(%
)
Silt Sand Gravel Cobble
261
5.4.4.2 Substrate fractions - Impacted sites
The substrate fractions in the impacted sites are outlined in Figs. 5.30 to
5.34. Boulders dominated the substratum in the Mad River and due to
difficulty in sampling it was therefore decided to omit these from the results.
Small cobble was recorded at all impacted sites while large cobble was
found only in the Cartron River and Lough na Corralea Stream (Appendix
5.16). Small cobble generally dominated the fractions in the impacted sites
followed closely by very coarse gravel. The Mullaghanoe River contained
the highest percentage of small cobble accounting for 62.99% of the total
fraction weight. Lough na Corralea stream contained the least amount
representing 22.69% of the total fraction weight. Large cobble represented
5.91% and 20.36% of the substrate fraction in the Cartron River and Lough
na Corralea Stream respectively. Very coarse gravel (32-64mm) was found
in all impacted sites ranging from 14.01% to 27.89% (Appendix 5.18).
262
Fig. 5.30 Mean frequency particle size distribution (by % weight) from the
Cartron River in March 2003. Error bars indicate maximum and minimum
values.
Fig. 5.31 Mean frequency particle size distribution (by % weight) from the
Lough na Corralea Stream in March 2003. Error bars indicate maximum and
minimum values.
Fig. 5.32 Mean frequency particle size distribution (by % weight) from the
Mad River in March 2003. Error bars indicate maximum and minimum
values.
0
10
20
30
40
50
60
70
80
90
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
12
8
25
6
Sieve size (mm)
Fre
qu
ency
(%
)
Silt Sand Gravel Cobble
0
10
20
30
40
50
60
70
80
90
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
128
256
Sieve size (mm)
Fre
qu
ency
(%
)
Silt Sand Gravel Cobble
0
10
20
30
40
50
60
70
80
90
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
12
8
25
6
Sieve size (mm)
Fre
qu
ency
(%
)
Silt Sand Gravel Cobble
263
Fig. 5.33 Mean frequency particle size distribution (by % weight) from the
Robe River in March 2003. Error bars indicate maximum and minimum
values.
Fig. 5.34 Mean frequency particle size distribution (by % weight) from the
Mullaghanoe River in March 2003. Error bars indicate maximum and
minimum values.
0
10
20
30
40
50
60
70
80
90
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
128
256
Sieve size (mm)F
req
uen
cy (
%)
Silt Sand Gravel Cobble
0
10
20
30
40
50
60
70
80
90
100
0.0
3
0.0
6
0.1
3
0.2
5
0.5 1 2 4 8
16
32
64
128
256
Sieve size (mm)
Fre
qu
ency
(%
)
Silt Sand Gravel Cobble
264
5.4.4.3 Cumulative frequency curves
The data from individual high status and impacted sites are presented in
cumulative frequency form in Appendices 5.19 and 5.20. Typically such
data yield an S-shaped curve. The main function of the curve is to show the
spread of particle sizes present. Mean values are used to report the results in
most cases.
These graphs show the broad range of particle sizes present at all sites as
can be observed by the flatness of the curve (i.e. tending to the horizontal).
This pattern is indicative of considerable homogeneity of substrate between
sites. Uniform samples would have a distribution that tends to the vertical.
There was a substantial variation in fraction size from different samples,
notably from one of the high status sites, the Owengarve River (Appendix
5.19A). This phenomenon was evident to a lesser extent in two of the
impacted sites i.e. the Cartron River (Appendix 5.20A) and the Mullaghanoe
River (Appendix 5.20E). At other sites, variation was considerably less.
For example, three very similar samples were recorded from the Brusna
River (Appendix 5.19E) the Mad River (Appendix 5.20C).
265
5.4.4.4 Fine sediments (particles < 1mm diameter)
Mean percentage fines were determined for each sample (Table 5.13).
Mean percentage fines in the high status rivers ranged from 0.46 to 15.33.
Mean percentage fines in the impacted rivers ranged from 1.03 to 1.95. The
lowest maximum value of 0.33 for fines was recorded at the Brusna River.
The highest mean, minimum and maximum values for fines were found in
the Owengarve River (15.3, 0.36 and 4.87 respectively). Mean percentage
fines for the high status sites are compared with the impacted sites (Fig.
5.35). Maximum and minimum values are included for each site. The
Owengarve River (high status) displayed the highest recorded mean
percentage fine value among all ten river sites.
Table 5.13 Mean percentage fines (particles <1mm diameter) from samples
obtained from the high status and impacted river sites on the 21st and 27th
March 2003.
River Mean Minimum Maximum
High status sites
Owengarve 15.33 0.36 4.87
Dunneill 2.11 0.01 1.49
Castlebar 0.46 0.01 0.34
Callow Loughs Stream 0.67 0.01 0.99
Brusna 0.50 0.01 0.33
Impacted sites
Cartron 1.42 0.01 0.98
Lough na Corralea Stream 1.06 0.01 0.01
Mad 1.03 0.01 0.75
Robe 1.33 0.02 0.68
Mullaghanoe 1.95 0.12 1.19
266
Fig. 5.35 Comparison of mean percentage fines (all particles < 1mm) from
the high status and impacted rivers sampled in March 2003.
The box and whisker graph in Fig. 5.36 shows the relationship in the mean
percentage fines (particles < 1mm diameter) between the high status and
impacted sites. There was no significant difference in the mean percentage
fines between the high status and impacted sites (Fig. 5.36: p<0.79). The
difference in the mean percentage frequency of all substrate fractions
between the high status and impacted sites is illustrated in Fig. 5.37 with no
significant differences recorded (p<0.71). The mean percentage fines were
compared between all ten river sites showing a significant difference (Fig.
5.38; p<0.001).
0
5
10
15
20
25
Sites
% o
f to
tal
sam
ple
wei
gh
t
High status rivers Impacted rivers
267
Fig. 5.36 Box and whisker graph showing the mean percentage fines (all
particles < 1mm diameter) between the high status and impacted sites:
p<0.79.
Fig. 5.37 Box and Whisker graph comparing the percentage frequency of all
substrate fractions between the high status and impacted sites. p<0.71.
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
Mea
n P
erce
nta
ge
fines
(<
1m
m)
-3.0
-1.5
0.0
1.5
3.0
4.5
High status rivers Impacted rivers
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
Per
cen
tag
e o
f to
tal
sam
ple
wei
gh
t
-25
-15
-5
5
15
25
35
45
High status rivers Impacted rivers
268
Fig. 5.38 Box and Whisker graph comparing the percentage fines between
the individual river sites. p<0.001.
The aim of sampling the substrate in the ten rivers was to identify the
particle sizes present at the various sites with particular reference to finer
material. On a river by river basis the high status rivers contained a higher
percentage of large cobble compared to the impacted sites. Small cobble
was found in all ten rivers and showed slightly higher percentages in the
high status rivers with the exception of one or two rivers. Boulders were
only found in one of the nine rivers, namely the Dunneill River.
The sediment fractions in the ten rivers were all quite similar with the
exception of the Owengarve River (high status river), which contained more
fine silts and gravel than any other site. In general, cobble (large and small)
and gravels (very coarse and coarse) dominated at all rivers. Overall, there
were no significant differences in the sediment across the ten river sites.
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
P
erce
nta
ge
freq
uen
cy
-2
0
2
4
6
8
Ow
eng
arv
e
Cas
tleb
ar
Bru
sna
Du
nn
eill
Cal
low
Lo
ug
hs
Str
eam
Car
tro
n
Lo
ug
h n
a C
orr
alea
Ro
be
Mu
llag
han
oe
Mad
High status rivers Impacted rivers
269
5.5 Discussion
5.5.1 Introduction
The focus of this study was to establish information on the value of the
physico-chemical data in supporting the Q-value System. It examines the
chemical environment and biotic features like the macroinvertebrate
communities and also the physical aspect with particular emphasis on the
substratum type all associated with five high status and five impacted rivers
in the West of Ireland.
An important objective of this project was to obtain an improved
understanding of reference conditions as revealed by changes in the
chemical parameters, macroinvertebrate fauna and physical changes when a
river’s status begins to depart from its pristine state. The project attempts to
refine our knowledge of the ecology of Ecdyonurus in relation to a range of
potential controlling factors in the riverine environment. In particular, it is
necessary to understand what controls the disappearance of Ecdyonurus as
eutrophication and organic pollution impacts on a river site.
5.5.2 General discussion
At the onset of the project, the high status rivers were chosen on the basis of
the presence of Ecdyonurus and of good water quality as evidenced by
water chemistry and high Q-values. The impacted sites were also chosen on
the knowledge that at one point Ecdyonurus also survived in these rivers,
and for reasons unknown, no longer exist. Ecdyonurus is a good indicator
of pollution and the water chemistry results appear to support the
hypothesis. The results show that ammonia was generally higher in the
impacted rivers than in the high status rivers. As a consequence of
eutrophication, on occasion however, some of the high status rivers
displayed elevated DO levels, high unfiltered MRP and TO.N
concentrations. This was particularly evident in the Owengarve River
indicating that they were not all at reference condition as defined by the
Water Framework Directive. Nonetheless, they were judged to be still of
270
high status according to the normative definitions of High Status in the
Water Framework Directive.
The conservative indicators of water quality, chloride, colour and
temperature displayed no significant differences between the high status and
impacted rivers. There was a significant difference in pH, conductivity and
alkalinity between the high status and impacted sites but these are most
certainly due to different river typologies, particularly as determined by
catchment geology.
Another hypothesis examined was whether low N:P ratios occur more
commonly at high status sites than in more impacted sites. The phosphate
levels were relatively high in some of the high status rivers and thus they
may have been N-limited on occasion during the sampling period. Results
from this study show that the impacted sites were more P-limited during the
sampling period compared to the high status sites. Some of the high status
rivers were N-limited, particularly the Castlebar and the Dunneill River
which may have contributed to the increased levels of phosphate in these
systems from time to time. Comparison with a wider range of physico-
chemical samples taken in the West of Ireland shows that occasional N-
limitation occurs when phosphate concentrations are less than 40µg P/l
(approximately 4% of all samples analysed). The wider survey also showed
that such N-limitation is significantly more likely to occur in the summer
months (July to October).
The analysis of the sediments across the river sites revealed no significant
differences. The sediment fractions in the ten rivers were all quite similar
with the exception of the Owengarve River (high status) which contained
more fine silt than any other site. This river was dredged a year prior to the
commencement of the project so the consequence of this was still evident.
The chalky nature of the surrounding banks and bank trampling by cattle
upstream may also have contributed to the higher content of silt. It might
have been expected that one would find less Ecdyonurus numbers in this
situation but in fact it was the opposite as the Owengarve River contained
the highest density and largest sizes of this species among all high status
271
sites. This, however, also is borne out by separate observations of
Heptageniidae survival even in highly turbid waters caused by inorganic silt
from sand quarry washings (McGarrigle pers. comm.). In general, the
sediment size across the ten rivers did not appear to be a factor in
controlling the presence or absence of Ecdyonurus. While the sediment
particle size analysis did not reveal significant differences between the
impacted and high status rivers, subjective observations of the substratum
did suggest that the impacted rivers were generally more ‘silted’ than the
high status group.
Understanding the changes that occur in macroinvertebrate communities in
relation to pollution is a key issue for impact assessment. There have been
numerous methods developed to measure community changes or to assess
water quality including a variety of biotic and diversity indices and
multivariate approaches (Hellawell, 1978, 1986; Metcalfe, 1989; Mason,
1991; Cairns et al, 1993; Rosenberg and Resh, 1993; Norris and Norris,
1995) and have been measured using a variety of terms, including biomass,
species richness and species composition and density. However, as Boyle et
al. (1990) pointed out, community changes are a complex function of
species composition, species richness and the relative abundance and
density of individuals.
The present study analysed the community differences within high status
and impacted sites in conjunction with an examination of various water
quality parameters. Several numerical indices were chosen in this study for
analysis and interpretation of the macroinvertebrate fauna. The ability of
the indices and metrics to highlight stressed sites varied. The diversity
indices were examined on a river by river basis across the sampling period
and the most significant differences between the high status and impacted
rivers were found using Margalef’s index, total number of taxa and % EPT.
All three indices/metrics were higher in the high status rivers. These
findings, therefore, also support the hypothesis that Ecdyonurus is a good
indicator of water pollution at the community level. The variation in the
diversity indices and biotic metrics was also examined across five seasons.
Contrary to expectations, there does not appear to be any difference in
272
sampling across the seasons (Appendix 5.5 to 5.11). On a river by river
basis one may see a change but there was not a marked, overall, seasonal
change in the metrics used to describe macrinvertebrate communities. This
may perhaps be due to the mild Irish climate which may produce more
gradual changes in community structure across the seasons than might be
expected in ecoregions where climate extremes are more pronounced.
Ecdyonurus was absent from three of the impacted rivers and was present
only on occasion in the other two sites, the Robe and the Mullaghanoe
rivers, which were both affected by eutrophication intermittently. These
findings support the accepted view that sensitive species are reduced when
water quality deteriorates (Chandler, 1970; Washington, 1984; Hellawell,
1986). However, it was also evident that polluted sites, particularly the
intermediately polluted sites like the Robe and the Mullaghanoe Rivers had
the highest abundance of invertebrates indicating an increase or influx of
benthic-invertebrates more tolerant to pollution.
Use of the recently available AQEM software to calculate a much wider
range of biotic indices including saprobic indices and modern European
indices such as the Danish Fauna Index also demonstrated significant
differences between impacted against high status sites. Trophic metrics like
feeding measures are surrogates of complex processes (e.g. trophic
interaction, production and food source availability) (AQEM consortium,
2002). Generalists, like gatherers and collectors have a broader range of
acceptable food materials than the specialists (e.g. grazers and scrapers) and
thus are more tolerant to pollution, which might alter availability of certain
food types. This may explain why the percentage of grazers and scrapers
was significantly higher (p<0.001) in the cleaner sites where foods ingested
by these taxa are more widely available. The metric results also show the
effect of eutrophication and organic pollution on the distribution of
functional feeding groups. The percentage of the gatherers/collectors, filter
feeders, predators, miners and parasites are significantly increased in the
impaired sites. It appears that food resources for the more specialised
grazers and scrapers may be more limited in these particular sites even
273
though they are only moderately polluted. This is confounded by the results
from the microhabitat preferences produced from the AQEM programme.
A significantly higher percentage of the macroinvertebrate fauna in the
higher status sites appeared to favour the Lithal microhabitat type (coarse
gravel, stones, boulders) indicating that the food ingested by these taxa are
more widely available on these substrate types. Results from AQEM show
that taxa like the sensitive indicator Ecdyonurus favoured these substrate
types suggesting that the surfaces contain good quality food sources for
these grazers. In comparison, a higher percentage of taxa in the impaired
sites favoured the Pelal microhabitat (mud loving) and the Akal
microhabitat (fine to medium-sized gravel). Even though the sediment
analysis did not show significant differences between the high status and the
impacted sites, the results from the feeding and microhabitat investigations
suggest a change in feeding guilds and microhabitat preferences among the
macroinvertebrate community as organic pollution and eutrophication
progresses.
Selection of the most appropriate indices and metrics were based primarily
on the ability of the index/metric to provide a meaningful summary of the
data as well as the applicability of the index to categorise the benthic
community. Based on findings from this study, Margalef’s index (d), total
number of taxa (S) and % EPT were the most appropriate metrics and
indices for distinguishing between high status and impacted sites in this
study. Biotic indexes used in other European countries which focus mainly
on organic pollution and eutrophication also gave clear differentiation
between the two groups of sites based on their macroinvertebrate
communities.
Studies carried out in the White River near Indianapolis investigated which
diversity, similarity and biotic indices best reflected the observed changes in
the water quality due to the effects of organic pollution (Lydy et al., 2000).
Results showed that the most descriptive tool in analysing the data was not
one of the indices but in fact the percentage EPT found at the site in a given
year. The trends shown by this tool best reflected the changes in water
274
quality at the three sampling sites. The diversity indices tested were the
least useful of all the indices; in fact, they indicated that the opposite result
had occurred and water quality declined after the wastewater treatment plant
located upstream of the site was upgraded (assuming high diversity
corresponds to better water quality). Diversity indices failed to reflect the
changes in the White River because the changes occurred in benthic-
invertebrate community structure more so than in diversity. Pollution-
intolerant EPT species replaced pollution-tolerant chironomids however,
and this observation was not incorporated or detected by the diversity
indices.
On the other hand, other studies have shown that taxon richness is a
powerful tool in distinguishing between non or slightly impaired and
stressed sites. Ofenbock et al. (2004) investigated the effects of various
types of pollution (including organic pollution) on a number of rivers in four
different bioregions in Austria. Richness measures reflect the diversity of
an assemblage and are known to be most useful in indicating impairment
(Resh et al., 1995). Elimination of taxa from a naturally diverse system can
be easily detected (Barbour et al., 1996). In particular, species belonging to
the insect orders Ephemeroptera, Plecoptera, and Trichoptera (EPT) are
generally regarded as sensitive to impairment and the loss of taxa richness
within this group indicates perturbation (Wallace et al., 1996). Only metrics
representing taxa richness and diversity (e.g. % EPT, total number of taxa,
intolerant taxa, diversity indices) showed a unidirectional response to
impacts like organic pollution. All were higher in the unimpacted rivers.
Margalef’s diversity index (Margalef, 1958) gave reliable results in
separating the stream types and this measure contributed to the indices of all
stream types with the exception of one. Unimpaired sites were more species
rich than the impaired ones.
Although the number of sites investigated in this study was comparatively
small, the findings are supported by other recent studies in highlighting the
high discriminatory power of Margalef’s index, total number of taxa and %
EPT in differentiating between impacted and unimpacted systems. These
may be very useful indicators of ecosystem health when assessing the
275
biological status of Irish rivers as demanded by the Water Framework
Directive. The Water Framework Directive defines high ecological status
and reference conditions very strictly. This project, however, was designed
to investigate the subtle changes a river experiences as it departs from a its
high status during the initial stages of eutrophication and the results may be
useful in providing input into the definition of the Good/Moderate boundary
for Irish rivers.
276
Chapter 6
6.1 General Discussion
6.1.1 Split-stream experiment
An important objective of this project was to obtain an improved
understanding of why ecological communities depart from reference
conditions as pollution and eutrophication impacts on individual species and
in particular indicator taxa such as Ecdyonurus. The effects of
eutrophication on Ecdyonurus were studied using a novel split-stream
experiment which involved artificially increasing the phosphorus
concentrations in two oligotrophic rivers in the West of Ireland. Nutrient
enrichment experiments usually involve the use of artificial enclosures or
channels in lakes and rivers but this experiment, the first of its kind
undertaken in Ireland, involved manipulating a river in its natural state.
Whole ecosystem enrichment experiments in lakes were pioneered by
Schindler and his fellow workers (Schindler, 1985) at the experimental lakes
in Canada. Similar experiments in rivers using artificial channels studying
enrichment or eutrophication have shown increases in periphytic biomass
under certain conditions (e.g. Elwood et al., 1981; Stockner and Shortreed,
1976; Horner and Welch, 1981; Horner et al., 1983; Perrin et al., 1987;
Bourassa and Cattaneo, 2000). Peterson et al., (1985) showed that an
increase of only 10µg/l PO4-P above the background (1-4µg/l P) in river
water channels, either alone or in combination with NO3-N, resulted in
substantial increases in epilithic chlorophyll a and other biological
activities. Other workers (MacKenthun, 1968; Wong and Clark, 1976;
Horner et al., 1983) involved in nutrient manipulation experiments in rivers,
found that much higher levels of phosphorus were needed to increase algal
biomass.
The first hypothesis examined in this investigation was whether Ecdyonurus
was directly or indirectly affected by the limiting nutrients phosphorus and
nitrogen. It was envisaged that by artificially enriching a P limited river,
277
algal growth would increase and the effects on the sensitive indicator genus
Ecdyonurus could be described. These experiments revealed surprising
results showing the importance of N-limitation in the rivers studied. Some
of the nutrient manipulation experiments showed significant differences in
algal biomass nonetheless between the control and treated sections, but not
all did so.
The results underline that experimental response is also dependent on the
N:P ratios and the temporal scale of the experiment. Due to a low N:P ratio
in the Clydagh River in 2002, no effect was observed in the manipulated
section after sampling on four occasions over a 6 week period. There was a
positive effect on periphyton growth in the same river in 2003 towards the
end of the 5-week experiment, even though the river apparently had a low
N:P ratio. The most significant differences in periphyton Chl a between
treated and untreated sections were observed in the Castlebar River in 2003,
particularly in the last 3 weeks of the 9-week experiment, despite an
apparently N-limited system.
There is the possibility that there was fluxing or pulsing of N through the
Clydagh and the Castlebar River systems, which was not detected between
sampling periods. Domestic houses and a concrete manufacturing facility
located upstream of the Castlebar River may have been potential sources
possibly introducing N sporadically into the river and particularly from
septic tanks of houses that are only occupied in the evenings and overnight –
i.e. outside times when grab samples from the river for water chemistry
purposes were taken. These undetected nitrogen sources may have caused
the fluctuating N:P ratios between sampling periods.
Experiments carried out by Bourassa and Cattaneo (2000) in an
experimental lake in Canada, showed that despite a four-fold increase in
phosphorus concentration, the periphyton Chl a was only slightly higher in
the enriched than the non-enriched treatments. Significant relationships
between periphytic chlorophyll and nutrients have often been observed
(Aizaki and Sakamoto, 1988; Biggs and Close, 1989; Mundie et al., 1991;
Lohman et al., 1992; Dodds et al., 1997; Harvey et al., 1998). When
278
nutrients are undetectable, loss processes like grazing or sloughing have
been proposed to explain the lack of response (Jones et al., 1984; Welch et
al., 1988; Kjeldsen, 1996; Bourassa and Cattaneo, 1998).
An increase in nutrients or light can augment invertebrate growth, density
and biomass (Lamberti et al., 1989; Mundie et al., 1991; Hill et al., 1995:
Dube et al., 1997; Bourassa and Cattaneo, 1998). In the present experiment,
no significant differences in numbers of Ecdyonurus were found between
the treated and untreated sections. These findings appear to indicate that
Ecdyonurus is not directly affected by the consequences of eutrophication
i.e. increased algal biomass at the relatively low levels of impact that were
observed during the experiment at any rate. It is hypothesised that nitrogen
spikes may have occurred that allowed available phosphate to be used thus
apparently causing the increased algal growth in some of the experiments,
particularly in the Clydagh River in 2003 and the Castlebar River in 2003
but having no direct effect on the abundance of Ecdyonurus.
On analysing the N:P ratios in 99 rivers in the West of Ireland it was found
that approximately 4% of the samples were N-limited with low MRP
concentrations (<0.05mg/l P). This implies that a small proportion of high
status rivers are nitrogen limited rather than phosphorus limited. Thus, in
terms of the Water Framework Directive, there may be a case for
introducing tighter regulations on the limits of nitrogen emitted to water
bodies as well as phosphorus. Results from these studies highlight the
complexity of the in-stream processes driving these N-limited and P-limited
high status rivers. It is clear that they are dynamic systems in a constant
state of flux.
6.1.1.1 Comments and recommendations
Nitrogen limitation appears to be a summer phenomenon primarily but more
detailed studies are required preferably over the course of an entire year, in
conjunction with investigations into the effect of nutrient enrichment in the
form of nitrogen on periphyton biomass growth and the associated
responses of sensitive indicator taxa.
279
6.1.2 Investigation into the feeding regime of Ecdyonurus venosus
It was hypothesised that the changes exhibited in the biomass in the split-
stream experiment would be reflected in the gut contents of Ecdyonurus
venosus, possibly showing a variation in algal taxa between both sections of
the manipulation experiment. This study provided an opportunity to assess
whether such direct manipulation had any obvious impact on the diet of this
species. The most significant differences between the enriched and
unenriched sections were observed in the Castlebar River in 2003 therefore
Ecdyonurus specimens sampled from this river were chosen for the main gut
analysis investigation. Studies dealing with algal-grazer interactions within
the mayfly community are under represented in the literature (Feminella and
Hawkins, 1995; Steinman, 1996), especially regarding the influence of
abiotic parameters on stream herbivory. Since the feeding preferences of
Ecdyonurus venosus had not been studied in Ireland to date, this study also
provided new information on its diet.
The hydrogen peroxide oxidation technique was not very successful in
isolating the benthic diatoms from the specimens and did not provide
information on the overall material consumed by these invertebrates. The
fluorochromatic stain 4’6 diamidino-2-phenylin-dole (DAPI) and
epifluorescent microscopy (Walker et al., 1988), in combination with light
microscopy, did however successfully categorise the material consumed in
detail. This provided much needed information into the classification of
Ecdyonurus into its functional feeding group by documenting its preferred
diet. Unfortunately, time constraints did not allow for any isotope analysis
on the gut contents, so this cheaper and less time consuming method of
analysing the gut contents provided new information into the diet of
Ecdyonurus. It paves the way for future work in gut analysis in Ireland as
the techniques for isolating the gut contents of Ecdyonurus are perfected and
further works are carried out in this area.
The gut contents of Ecdyonurus venosus were examined over a number of
sampling occasions and preliminary findings reveal promising results in
relation to what these invertebrates feed on. The study documented the
280
material consumed but did not provide information on what was being
assimilated so the nutritional significance of the ingested food or the manner
in which they fed were not examined.
On the basis of the findings from the study in the Castlebar River in 2003,
Ecdyonurus venosus can be classified as both a herbivorous grazer and
detritus feeder with a tendency towards opportunistic feeding. Moog (1995)
also proposes that this genus feeds in a dual action mode and describes it as
both a grazer and a detritus feeder. The food ingested by these larvae
appears to be strongly dependent on the food available in the environment at
a given time and they seem to feed on particles that are most abundant
during a particular season or those that are easily accessible during a feeding
episode. This may explain the differences in the proportions of diatoms in
the invertebrate guts investigated in March 2003 compared to those studied
from July-September 2003.
The Ecydonurus gut contents consisted mainly of epilithic algal tissue, plant
particulate matter (detritus), biofilm matrix and inorganic debris (mineral
material). Results indicate a greater proportion of inorganic debris in the
gut of Ecdyonurus in comparison with that found on the stone scrapings
more than likely due to the brushing action of Ecdyonurus and its tendency
to harvest overstorey layers. Studies carried out by Wellintz and Ward
(1998) indicate that Ecdyonurus also harvested overstorey layers. Their
findings suggest that most overstorey periphyton removed by Ecdyonurus
appeared to be an amalgam of diatoms, silt and detritus. The most common
algal species found in the guts of the invertebrates sampled in the Castlebar
River from July-September 2003 was Navicula spp. The diet seems to
depend on the presence of a particular diatom species in the habitat in any
given season and preference to a particular species may be due to a greater
ease in scraping some diatom species off the substratum than others.
Interaction between periphyton and grazers can be confounded by
contrasting impacts for grazers and differential algal responses to grazing
(Pan and Lowe, 1994). Periphytic algal biomass was stimulated in the final
weeks of the experiments in 2003 (particularly in Castlebar River) and it
281
was hypothesised that there may be a change in the algal taxa between the
control and treated sections of the split-stream experiment. As mentioned in
Chapter 3, Paul and Duthie (1989) have shown that algal species respond to
increases in phosphorus by changing their community structure. Similar
community shifts resulting from nutrient manipulation was also reported by
Pringle and Bowers (1984). This was not evident in this investigation,
however, which was reflected in the similarity in the stone scrapings and the
food content in the guts of the larvae examined on both sides of the
experimental divide. The hypotheses that Ecdyonurus would demonstrate
diet changes due to enrichment was not supported but neither were there any
observed changes in the periphyton species community.
6.1.2.1 Comments and recommendations
An in-depth study, preferably over a 12-month period across seasons is
essential in this area in order to provide a more complete understanding into
the feeding regime of Ecdyonurus. It should also be possible to determine
which foods are assimilated using stable isotope analysis. In terms of
understanding the role of diet in the known sensitivity of Ecdyonurus to
pollution and eutrophication this study has shown that it has a relatively
specialised mode of feeding. It is suggested that as eutrophication
progresses and filamentous algae such as Cladophora begin to blanket stone
surfaces it will become increasingly difficult for Ecdyonurus to feed in its
normal grazer mode.
282
6.1.3 Life history studies of the Heptageniidae in five high status
rivers in the West of Ireland
To understand and predict the response of organisms to variation and
change within and between lotic ecosystems, we need information about
their life histories (Power et al., 1988). The life cycle of the genus
Ecdyonurus has not been described in any great detail in Ireland to date
therefore this investigation provided much needed new information on this
topic.
Of the three species studied, Ecdyonurus venosus was the dominant species
in all five high status rivers displaying a bivoltine life cycle with only a
slight variation in emergence periods between sites. Conflicting findings in
Ireland were documented by Connolly and McCarthy (1993) and they
described its life cycle as being univoltine in the Corrib catchment. As in
this study, Fahy (1973) describes Ecdyonurus venosus as having a bivoltine
life cycle. It may be that the lifecycle is flexible, having a bivoltine or a
univoltine cycle depending on the temperature regime of individual rivers or
in a given year. Studies carried out in England by Elliott (1967) and Wise
(1980) show that this species had a univoltine life cycle which disagrees
with the findings of Rawlinson (1939) who found that it had a fast growing
summer generation in addition to a slow growing winter generation
(bivoltine).
The life cycle of Ecdyonurus dispar investigated in this study was
univoltine. Similar findings were documented in English studies carried out
by Macan and Maudsley (1968) and Wise (1980). There was no evidence in
the literature relating to studies on the life cycle of Ecdyonurus dispar or
Ecdyonurus insignis in Ireland. Ecdyonurus insignis had a univoltine
lifecycle in the rivers studied in this investigation. Macan (1970) also
described it as having a similar life cycle.
The absence of both Ecdyonurus insignis and Ecdyonurus dispar from the
benthos during the summer and winter months suggest that these species
develop quite differently to the more common species Ecdyonurus venosus.
283
Partitioning of emergence periods appears to be evident in this study and is
particularly noticeable in the Owengarve River where all three species were
present. Macan (1981) describes this phenomenon as a tool that may ensure
the survival of the next generation of each species. Other studies described
by Landa (1968) and Sowa (1975) underpins the suggestion from our
findings that the different species in the same river follow chronological
patterns of emergence periods maximising the survival of each individual
species. Lotic species are often regionally different in age and size at first
reproduction, in numbers of generations per year and in the degree of
synchrony of life history stages (Newell and Minshall, 1978).
The life cycle of Rhithrogena semicolorata was more straightforward and
easier to interpret and it clearly had a univoltine life cycle. Studies carried
out by Wise (1980) and Elliott and Humpesch (1980) in England describe it
as being univoltine. The only study relating to its life cycle in Ireland was
carried out by Fahy in 1971, where he also describes it as having a
univoltine life cycle. The Heptagenia specimens were identified to genus
level only and were therefore described as a genus group that appeared to
adopt a univoltine life cycle with an overwintering larval generation. It was
the least common of the Heptageniidae family and apart from investigations
carried by Wise (1980) in England, there were no other similar studies
found in the literature.
Apart from being absent from the benthos for a few months after the main
flight period, particularly in July and August, Rhithrogena semicolorata was
also present throughout the year, so one would therefore expect to find this
species during most seasons in a given year also. Its absence during July
and August, however, does not have the same significance as does the
absence of Ecdyonurus during the summer months. Hepatgenia spp.
emerged a bit later than Rhithrogena semicolorata so depending on the
abundance in a river, would be expected to be found throughout the year
when routinely sampling, apart from July and August and in some rivers in
September.
284
6.1.3.1 Comments and recommendations
In order to ensure that the Heptageniidae, and in particular the genus
Ecdyonurus, are fully represented when carrying out routine biological
assessments, the timing of sampling may be a critical factor, especially
when attempting to capture species that are only present in the benthos for a
few months during the year. Findings from the present studies support the
hypothesis put forward that the various species of Ecdyonurus emerge in
overlapping phases such that during the summer months larvae of at least
one Ecdyonurus species will be present in the benthic riffle fauna of Irish
rivers. Results indicate that more intense sampling may be required during
certain months of the summer.
Thus, assuming adequate sampling, the overlapping life cycles of
Ecdyonurus should always result in summer samples yielding at least one
representative of Ecdyonurus if a river is of high status or at reference
condition. This is important if the monitoring programme is concerned with
water quality as opposed to biodiversity or taxonomic issues. Summer low
flow and high temperature conditions will tend to aggravate the impact of
pollution discharges and thus, the faunal community in summer and early
autumn provide a ‘minimum thermometer’ measure of the impact of
pollution. This is the time when fish kills or loss of sensitive
macroinvertebrates is most likely to occur. While many species have life
cycles that help them to avoid critical conditions the present study suggests
that at least one species of Ecdyonurus is likely to be present in high status
rivers throughout the summer months.
Autumn or winter sampling may produce more species both of Ecdyonurus
and other pollution sensitive taxa even in polluted systems as aestivating
eggs of plecopteran species and Rhithrogena hatch out and small nymphs
begin to appear in macroinvertebrate samples. If the aim of the sampling,
however, is to assess water quality or ecological status as impacted by
anthropogenic effects in particular, then summer assessments provide a
more reliable assessment of worst case conditions in a river. As it may only
take one pollution event to severely alter the taxonomic composition of a
285
lotic community this is a critical point. Autumn or winter sampling may not
be sufficiently sensitive to detect for example summer eutrophication effects
such as low night time dissolved oxygen during warm low flow conditions.
From studying the life cycles of the Heptageniidae in five rivers in the West
of Ireland it appears that the life cycles of the genus Ecdyonurus can vary
slightly from year to year and from river to river. Due to the short life cycle
of Ecdyonurus insignis and in particular Ecdyonurus dispar, which appear
to be summer species, it would be vital to sample from June to August in a
given year to ensure capture of these species. The more common species of
this genus, Ecdyonurus venosus was present in each of the rivers studied
throughout the year, with the exception of a few weeks post emergence.
Hence, one would expect to find this species when sampling during all
seasons throughout the year. It reinforces the indicator value of
Ecdyonurus.
286
6.1.4 Examination of the biotic and abiotic factors controlling the
distribution of the genus Ecdyonurus
There are many possible indicators of river health, including measures of
structure and function both of the biotic and of the physical components.
Physical and chemical indicators (mostly of water quality) are the most
commonly used and largest variety available (e.g. Hart et al., 1999; Maher
et al., 1999). One of the main objectives of this study is to understand what
controls the disappearance of Ecdyonurus as eutrophication and organic
pollution impact on a river. The occurrence of Ecdyonurus is controlled by
a number of features including chemical, physical and biotic factors.
Measuring the health of a river system should therefore include an
assessment of the biological community and its physico-chemical
characteristics.
Ecdyonurus is a good indicator of pollution and the water chemistry results
appear to support the hypothesis that the presence of Ecdyonurus is
associated with good water quality. Ammonia concentrations were
generally higher in the impacted rivers than in the high status rivers. As a
consequence of eutrophication, on occasions, some of the high status rivers
displayed elevated DO levels, high unfiltered MRP and TON
concentrations.
There were no significant differences in the conservative indicators of water
quality, namely chloride, colour and temperature between the high status
and impacted rivers. As mentioned in Chapter 5, significant differences in
pH, conductivity and alkalinity between the high status and impacted sites
were most certainly due to different river typologies, which were largely
dependent on the catchment geology.
The presence/absence of Ecdyonurus in the high status v impacted sites
supports its use as a significant bioindicator of water quality. The presence
of Ecdyonurus did not appear to be controlled by any of the physico-
chemical parameters examined in this experiment. This genus did survive
on occasion in some of the impacted sites that were moderately polluted
indicating that favourable conditions were present from time to time to
287
support their survival. Despite the fact that the ammonia levels were
significantly higher in the impacted sites, Ecdyonurus was evident in a
number of these rivers.
No significant differences were found in the sediments across the ten river
sites. It was evident that some of the sites were affected by siltation during
the sampling programme, in particular the Owengarve and the Mullaghanoe
Rivers. The sampling technique applied to the sediment analysis did
emphasise the large amount of fine silt in the Owengarve River but it did
not detect this in the Mullaghanoe or the Robe River. The impacted rivers
appeared to be more ‘silted’ than the high status rivers from time to time
highlighting the inadequacy of sampling on just one occasion. The results
suggest that traditional particle size analysis may not be sensitive enough to
detect the changes in habitat that have occurred but that the known
microhabitat preferences of the invertebrate community taxa may allow
quite subtle changes to be detected. It is hypothesised that the condition and
precise nature of the surface films on stones is perhaps more important than
the absolute particle size distribution of the gross substratum samples that
were taken.
A number of numerical indices and metrics were selected to analyse and
interpret the macroinvertebrate communities in the high status and impacted
sites. The ability of the indices and metrics to emphasise stressed sites
varied and the results revealed that the most significant differences between
the two groups of sites were found using Margalef’s index, total number of
taxa and % EPT. All three indices were higher in the high status rivers. A
wider range of biotic indices became available from the recent introduction
of the AQEM system also showing significant differences between the high
status and impacted sites.
The most appropriate indices were chosen on their capacity and suitability
for assessing the impact of organic pollution and eutrophication on the
bentic faunal community among the two groups of sites. Results from the
feeding and microhabitat indices used in AQEM suggest that as
eutrophication and the impacts of organic pollution progress, a change in the
288
feeding guilds and microhabitat preferences among the macroinvertebrate
communities occur.
It has been observed that increased sediment loads may reduce the
abundance and diversity of invertebrates by smothering interstitial habitat
and reducing periphytic abundance or quality (Lloyd et al., 1987;
Newcombe and McDonald, 1991; Ryan, 1991; Wood and Armitage, 1997).
Fine silts have been found to be unsuitable habitats for most New Zealand
aquatic insects (Quinn and Hickey, 1990; Jowett et al., 1991; Death, 2000).
Some New Zealand invertebrate species, notably Deleatidium spp.
(Ephemeroptera, Leptophlebiidae) and Pycnocentrodes sp. (Trichoptera,
Conoesucidae) show preferences for ‘clean’ rather than silted periphyton
(Ryan, 1991) and in colonisation trials, Ryder (1989) showed that the
occurrence of fine sediment in the algal matrix reduced invertebrate
densities by about 30%.
Consequently, changes in the pattern of sediment deposition as a result of
landuse change may modify the impact that invertebrate grazers have upon
periphytic prey. The Hydrobiid snail Potamopyrgus antipodarum for
example, is often dominant in the macroinvertebrate communities in low-
order pasture streams throughout New Zealand, yet it is rare in otherwise
similar afforested catchments in which the ‘sensitive’ ephemeropteran,
plecopteran and trichopteran taxa usually dominate (Quinn and Hickey,
1990). Of these latter taxa, the Leptophlebiid mayfly Deleatidium sp. is
often the most abundant species and this has been shown to be unable to
separate food from silt prior to ingestion and to be less abundant on
sediment-rich epilithic tiles than on sediment poor ones (Ryder, 1989; Ryan,
1991). In contrast, P.antipodarum was more abundant on the impacted tiles
(Ryder, 1989) and has been shown to take (small) sediment particles into the
buccal cavity, scrape the encrusting organic matter off and finally ‘spit’ the
sediment particle out (Lopez and Kofoed, 1980). These various lines of
evidence suggest that P.antipodarum may be more tolerant of sediment
contamination in its food than Deleatidium sp. are.
289
It is surmised that the changes in microhabitat may predominantly be
affecting surface films. As eutrophication progresses the nature of surface
films on stones and in the interstices of riverine gravels may change. Thus,
increases in absolute abundance and production of surface algae – diatoms,
cyanobacteria and green algae, for example. Changes in algal species may
occur. Based purely on subjective observation of surface films on stones, it
may also by hypothesised that increased pick-up of inorganic silt and
organic detritus is occurring. While the initial hypothesis suggested that
particle size analysis would be sufficient to detect ongoing siltation effects it
now appears that at the levels of deterioration experienced at the impacted
sites sampled, this analysis was not sufficiently sensitive. It should be borne
in mind that Irish catchments (and particularly in the West of Ireland) have
very little tillage agriculture and thus, significant silt inputs of soil origin are
not expected. Peat silt due to exploitation of bogs, forestry drainage or due
to overgrazing of blanket bogs may, however, be quite significant in certain
areas although the Mad River is likely to be the only site so affected in the
impacted sites. Future work should examine changes in surface films in
more detail in an attempt to support or refute the above hypothesis.
Food quality can influence key life history traits of aquatic insects such as
growth rate, size at maturity, and the ability to complete metamorphosis and
to reproduce (Anderson and Cummins, 1979; Cargill et al., 1985; Dadd,
1982; Lamberti and Moore, 1984). Several studies have shown that because
of the low quality of detritus as food, it supports only a small portion of the
growth of hydropsychid caddisflies despite its abundance in the insects guts
(Beneke and Wallace, 1980; Fuller and Mackay, 1981; Haefner and
Wallace, 1981). The nature and consequently the digestibility of the
detritus, algal tissue, biofilm matrix (organic film on rocks) and other
material may be different from that consumed by other insects. In
establishing stability within the invertebrate community in rivers it is
important therefore that each taxon has a good supply of its own required
food sources. High status rivers display a balanced ecosystem where food is
plentiful and grazing pressures are able to sustain an equilibrium with
periphyton growth.
290
6.1.4.1 Comments and recommendations
The analyses support the findings of the split-stream study, suggesting that
nitrogen limitation may be an important aspect of eutrophication and of
reference conditions in certain river types and particularly during the
summer months. More work is required to elucidate the reasons why a
small number of high status sites are nitrogen limited rather than the more
typical state of phosphorus limitation.
The results of the whole-community analysis give some potential insights
into the manner in which the traditional biotic indexes of organic pollution
and eutrophication work at a community level. The traditional pollution-
sensitive indicator species have definite microhabitat and feeding
requirements that are typically found in oligotrophic unpolluted rivers. As
nutrient levels increase the microhabitat changes sufficiently to favour less
sensitive species. At the extreme an anoxic, mud-dominated environment
will favour only tubificid worms and Chironomus perhaps, but there is a
continuum of change and microhabitat effects may be more important than
simple deoxygenation effects or toxicity effects due to ammonium, for
example, at the early stages of eutrophication.
Development of alternative sampling methodologies is required to measure
the effects of siltation that were undetected by the technique used in this
investigation. Further studies into the microhabitat and feeding preferences
of Ecdyonurus are essential in conjunction with a more detailed examination
into the changes in the food quality that occur as eutrophication and the
effects of organic pollution progress.
291
6.1.5 Concluding comments
The results demonstrate that the factors controlling biological communities
may be complex both in time and space. The limited nocturnal oxygen
survey carried out was perhaps insufficient to demonstrate the true impact of
low dissolved oxygen saturation in these rivers. Macroinvertebrates,
however, act as ‘minimum thermometers’ in the sense that they are
impacted by the worst conditions – e.g. the lowest oxygen or the highest
ammonia concentrations – that prevail while they are present in the river.
Ecdyonurus is known to be sensitive to low oxygen saturation and is
generally absent from moderately polluted river sites in Ireland.
The life cycle analyses in this study suggest that at least one species of
Ecdyonurus should be present at all times of the year. Thus, their utility as
an indicator species is justified in the sense that if they are absent at a site it
is most likely because of an adverse environmental pressure. It is postulated
that the nature of the nutrient rich organic biofilms that once covered the
stones in the impacted sites where Ecdyonurus previously survived have
been adversely affected. Findings from the gut analysis studies show that
epilithic algal tissue, plant particulate matter (detritus), biofilm matrix and
inorganic debris (mineral material) are the main food items ingested by
Ecdyonurus. The food quality of the biofilm, now deemed to be a very
important source of nutrients for macroinvertebrates, may be reduced in
these impacted sites due to adhesion of fine sediment particles undetected in
this study. Our studies showed that algae are an important food source in
the diet of Ecdyonurus and although not studied in detail in this
investigation, changes in the epilithic algae in the impacted sites may have
affected its feeding habits.
An understanding of the detailed ecology of key indicator species like
Ecdyonurus is important in defining and comprehending the ecology of high
status river sites. At high status sites the important indicator species are
able to flourish and maintain sustainable populations. Increased knowledge
of the detailed ecology of indicator species and reference conditions is
necessary in order to enable cross-European comparisons between different
292
ecoregions and different ecotypes. Findings from this study have increased
our knowledge in defining reference conditions, which is critical for the
development of classification systems. In terms of the WFD, the results are
beneficial and will be particularly useful when carrying out intercalibration
exercises with other European countries and with other ecoregions.
293
References
Agences de l’Eau (2000) System for the evaluation of the biological quality
of watercourses . SEQ Bio (Version 0). General principles. Water Agencies
Study No. 77.
Aizaki, M. and Sakamoto, K. (1988) Relationship between water quality
and periphyton biomass in several streams in Japan. Verhandlungen der
internationale Vereinigung fur theoretische und angewandte Limnologie,
23, 1511-1517.
Alba Tercedor, J. and Sanchez Ortega, A. (1988) Un metodo rapido y
simple para evaluar la calidad biologica de las aguas corrientes basado en el
de Hellawell (1978). Limnetica, 4, 51-56.
Allen, J.D. (1995) Stream ecology. Structure and function of running
waters. Chapman and Hall, London, 388pp.
Anderson, N.H. and Cummins, K.W. (1979) Influences of diet on the life
histories of aquatic insects. Journal of the Fisheries Board of Canada. 36,
335-342.
AQEM consortium (2002) Manual for the application of the AQEM system.
A comprehensive method to assess European streams using benthic
macroinvertebrates, developed for the purpose of the Water Framework
Directive. Version 1.0 February 2002, 198pp.
AQEM Version 2.3.4a August (2004) AQEM River Assessment
Programme. Developed under EU 5th
Framework Project Developing and
testing of and Integrated Assessment system for the Ecological Quality of
Streams and Rivers Throughout Europe using Benthic Macroinvertebrates.
Contract EVK1-CT1999-00027. Website: www.aqem.de
Arens, W. (1989) Comparative functional morphology of the mouthparts of
stream animals feeding on epilithic algae. (Monographische Beiträge). –
Archiv für Hydrobiologie, 83, 253–534.
Armitage, P.D., MacHale A.M. and Crisp, D.C. (1975) A survey of the
invertebrates of four streams in the Moor House National Nature Reserve in
North England. Freshwater Biology, 5, 479-495.
Austrian Standards ONORM M 6232 (1997) Guidelines for the Ecological
Survey and Evaluation of Flowing Surface Waters. Austrian Standards
Institute, Vienna, 38 pp.
Axler, R.P., Rose, C. and Tikkanen, C.A. (1994) Phytoplankton nutrient
deficiency as related to atmospheric nitrogen deposition in Northern
Minnesota acid-sensitive lakes. Canadian Journal of Fish and Aquatic
Sciences, 51, 1281-1296.
294
Barbour, M.T., Gerritsen, J., Griffith, G.E., Frydenborg, R. McCarron, E.,
White, J.S. and Bastian, M.L. (1996) A framework for biological criteria for
Florida streams using benthic macroinvertebrates. J.N. Am. Benthol.Soc. 15,
185-211.
Barbour, M.T., Gerritsen, J., Snyder, B.D. and Stribling, J.B. (1999)
Revision to Rapid Bioassessment Protocols for Use in Streams and Rivers:
Periphy..ton, Benthic Macroinvertebrates and Fish. US Environmental
Protection Agency Assessment and Watershed Protection Division,
Washington, DC. EPA 841-D-97-002.
Basu, B.K. and Pick, F.R. (1996) Factors regulating phytoplankton and
zooplankton biomass in temperate rivers. Limnol. Ocenaogr, 41, 1572-1577.
Bauernfeind, E. (1994) Bestimmungsschlussel fur die osterreichischen
Eintagsfliegen (Insecta: Ephemeroptera), 1. Teil. Wasser Abwasser Suppl.
4, 1-92.
Bauernfeind, E. (1997) Discriminating characters in Central European
species of Ecdyonurus EATON. In Landolt P. & M. Sartori (eds),
Ephemeroptera & Plecoptera: Biology-Ecology-Systematics (= Proc. 8th
Int.
Conf. Ephemeroptera, Lausanne, 1995), Fribourg, 418-426.
Bauernfeind, E. (1998) Erganzugen und Berichtigungen zu den
Eintagsfliegen des Lavanttales (Insecta: Ephemeroptera). Linzer. Biol. Beitr.
30, 293-298.
Bauernfeind, E. and Moog, O. (2000) Mayflies (Insecta: Ephemeroptera)
and the assessment of ecological integrity: a methodological approach.
Hydrobiologia. 422/423, 71-83.
Beneche, V. (1972) La fecondite be Baetis rhodani Pictet. Freshwater
Biology, 2, 337-354.
Benke, A.C. (1984) Secondary production of aquatic insects, pp. 289-322.
In: The Ecology of Aquatic Insects, V.H. Resh and D.M. Rosenberg (eds),
Praeger, New York.
Beneke, A.C. and Wallace, J.B. (1980) Trophic basis of production among
net-spinning caddisflies in a southern Appalachian stream. Ecology, 61,
108-118.
Biggs, B.J.F. and Close, M.E. (1989) Periphyton biomass dynamics in
gravel bed rivers: the relative effects of flow and nutrients. Freshwater
Biology, 22, 209-231.
Biggs, B.J.S. and Lowe, R.L. (1994) Response of two trophic levels to patch
enrichment along a New Zealand stream continuum. New Zealand
J.mar.freshwat.Res., 28, 119-134.
295
Biggs, B.J.F. and Thomsen, H.A. (1995) Disturbance of stream periphyton
to perturbation in shear stress: time to structural failure and differences in
community resistance. Journal of Phycology, 31, 233-241.
Birk, S. and Hering, D. (2002) Waterview Web-Database: a comprehensive
review of European assessment methods for rivers. FBA News 20 (winter
2002): 4.
Blinn, D.W., Fredericksen, A. and Korte, V. (1980) Colonization rates and
community structure of diatoms on three different rock substrata in lotic
system. British Phycological Journal, 15, 303-310.
Bohle, H.W. (1978) Beziehugen zwischen dem Nahrungsangebot, der Drift
und der raumlichen Verteilung bei Larven von Baetis rhodani (Pictet)
(Ephemeropteren: Baetidae) Untersuchungen in Kunstlichen
Fliesswasseranlagen. Archiv fur Hydrobiologie, 84, 500-525.
Borchardt, M.A. (1996) Nutrients. In: Algal Ecology: Freshwater Benthic
Ecosystems (eds R.J. Stevenson, M.L. Bothwell and R.L. Lowe), pp. 183-
227. Academic Press, San Diego, CA.
Bothwell, M.L. (1985) Phosphorus limitation of lotic periphyton growth
rates: an intersite comparison using continuous flow troughs. (Thompson
River System, British Columbia). Limnol. Oceanogr, 30, 527-542.
Bott, T.L. (1983) Primary production in streams. Stream Ecology:
Application and Testing of General Ecology Theory. (eds. J. R Barnes and
G.W. Minshall), pp. 29-54. Plenum Press, New York.
Bourassa, N. and Cattaneo, A. (1998) Control of periphyton biomass in
Laurentian streams. Journal of the North American Benthological Society,
17, 420-429.
Bourassa, N. and Cattaneo, A. (2000) Response of a lake outlet community
to light and nutrient manipulation: effects on periphyton and invertebrate
biomass and composition. Freshwater Biology, 44, 629-639.
Bowman, J.J. and Clabby, K.J (1998) Water quality of rivers and lakes in
the Republic of Ireland. Environmental Protection Agency. In: Wilson, J.G.
(ed.) 1998 Eutrophication in Irish waters, 55-63. Dublin. Royal Irish
Academy.
Boyle, T.P., Smillie, G.M., Anderson, J.C and Beeson, D.R. (1990) A
sensitive analysis of nine diversity and seven similarity indices. J. Wat.
Pollut. Control. Fed, 62, 749-762.
Brabec, K. Zahradkova, S., Nemejcova, D., Paril, P, Kokes, J. and
Jarkovsky, J. (2004) Assessment of organic pollution effect considering
differences between lotic and lentic stream habitats. Hydrobiologia, 516,
331-346.
296
Breitenmoser-Wursten, C. and Sartori, M. (1995) Distribution, diversity, life
cycle and growth of a mayfly community in a prealpine stream (Insecta,
Ephemeroptera). Hydrobiologia, 308, 85-101.
Brittain, J.E. (1982) Biology of mayflies. Annual Review of Entomology, 27,
119-47.
Brittain, J.E. (1991) Life history characteristics as a determinant of the
response of mayflies and stoneflies to man-made environmental disturbance
(Ephemeroptera and Plecoptera). Proc. Vlth. Int. Conf. Ephemeroptera and
Plecoptera, Granada: 539-545.
Brown, M.E. (1957) Experimental studies on growth. In: The physiology of
fishes (Ed. M. E. Brown), Vol. 1. pp.361-400. Academic Press London.
Brown, E.J. and Button, D.K. (1979) Phosphate-limited growth kinetics of
Selenastrum capricornutum (Chlorophyceae). J. Phycol, 15, 305-311.
Buffagni, A., Erba, S., Cazzola, M. and Kemp, J.L. (2004) The AQEM
multimetric system for the southern Italian Apennines: assessing the impact
of water quality and habitat degradation on pool macroinvertebrates in
Mediterranean rivers. Hydrobiologia, 516, 313-329.
Bunn, S.E., Edwards, D.H. and Lonergan, N.R. (1986) Spatial and temporal
variation in the macroinvertebrate fauna of streams of the northern jarrah
forest, Western Australia: community structure. Freshwater Biology, 16, 67-
91.
Butler, M.G. (1984) Life histories of aquatic insects. In The ecology of
aquatic insects (ed. V.H.Resh and D.M. Rosenberg), pp.24-55. Praeger,
New York.
Cairns, J. Jr, McCormick, P.V. and Niederlehner, B.R. (1993) A proposal
framework for developing indicators of ecosystem health. Hydrobiologia,
263, 1-44.
Canfield, D.E. Jr. (1983) Prediction of chlorophyll a concentrations in
Florida lakes: the importance of phosphorus and nitrogen. Wat. Res. Bull,.
19, 255-262.
Cargill, A.S., Cummins, K.W., Hanson, B.J. and Lowry, R.R. (1985) The
role of lipids, fungi and temperature in the nutrition of a shredder caddisfly,
Clistoronia magnifica. Freshwater Invertebrate Biology, 4, 64-78.
Cattaneo, A. and Kalff, J. (1986) The effects of grazer size manipulation on
periphyton communities. Oecologia, 69, 612-617.
Chandler, J.R. (1970) A biological approach to water quality management.
Wat. Pollut. Control London, 69, 415-422.
Chapman, D.W. and Demory, R.L. (1963) Seasonal changes in the food
ingested by aquatic insect larvae and nymphs in two Oregon streams.
Ecology, 44, 140-146.
297
Chovanec, A., Jager, P., Jungwirth, M., Koller-Kreimel, V., Moog, O.,
Muhar, S. and Schmutz, S. (2000) The Austrian way of assessing the
ecological integrity of running waters: A contribution to the EU Water
Framework Directive. Hydrobiologia, 422/423, 445-452.
Chutter, F.M. (1972) A reappraisal of Needham and Usinger’s data on the
variability of stream fauna when sampled with a Surber sampler. Limnology
ann Oceanography, 17, 139-141.
Clabby, K.J., Lucey, J, McGarrigle, M.L., Bowman, J.J., Flanagan, P.J. and
Toner, P.F. (1992) Water Quality in Ireland 1987-1990, Part One: General
Assessment. Water Resources Division, An Foras Forbartha, Dublin. 189pp.
ISBN 1 85053 180 3.
Clarke, G.M. (1994) Developmental stability analysis: an early-warning
system for biological monitoring of water quality. Australian Biologist, 7,
94-104.
Clarke, G.M., Arthington, A.H. and Pusey, B.J. (1995) Fluctuating
asymmetry of chironomid larvae as an indicator of pesticide contamination
in freshwater environments. Pages 101-109, In P.S. Cranston (ed.)
Chironomids: from genes to ecosystems. CSIRO, Melbourne.
Clenaghan, C., Giller, P.S., O’Halloran, J. and Hernan, R. (1998) Stream
macroinvertebrate communities in a conifer-afforested catchment in Ireland:
relationships to physico-chemical and biotic factors. Freshwater Biology,
40, 175-193.
Clifford, H.F. (1982) Life cycles of mayflies (Ephemeroptera), with special
reference to Voltism. Quaestiones Entomologicae, 18, 15-90.
Closs, G.P. and Lake, P.S. (1994). Spatial and temporal variation in the
structure of an intermittent-stream food web. Ecological Monographs, 64,
1-21.
Coates, D. (1980) Prey-size intake in humbug damselfish, Dascyllus
aruanus (Pisces, Pomacentridae) living within social groups. Journal of
Animal Ecology, 49, 335-340.
Coffman, W.P., Cummins, K.W. and Wuycheck, J.C. (1971) Energy flow
in a woodland stream ecosystem: I. Tissue support structure of the autumnal
community. Archiv fuer Hydrobiologie, 68, 232-276.
Coleman, A.W. (1980) Enhanced detection of bacteria in natural
environments by fluorochrome staining of DNA. Limnol. Oceanogr. 25,
948-951.
Coleman, M.J. and Hynes, H.B.N. (1970) The life histories of some
Plecoptera and Ephemeroptera in a southern Ontario stream. Can. J. Zool.
48, 1333-1339.
298
Colletti, P.J., Blinn, D.W., Pickart, A and Wagner, V.T. (1987) Influence of
different densities of the mayfly grazer Heptagenia criddlei on lotic diatom
communities. J. North. Am. Benthol. Soc. 6, 270-280.
Connolly, B. and McCarthy, T.K. (1993) Maflies (Ephemeroptera) of the
Corrib catchment, Ireland. Biogeography of Ireland: past, present and
future. M.J. Costello and K.S. Kelly (eds.). Occ. Publ. Ir. Biogeog. Soc. No
2.
Cooke, G.D., Welch, E.B., Peterson S.A. and Newroth, P.R. (1993)
Restroation and Management of Lakes and Reservoirs. 2nd
edn, 548 p.
Lewis Publishers, Boca Raton.
Cox, E.J., (1990) Studies on the algae of a small softwater stream I.
Occurrence and distribution with particular reference to the diatoms. Archiv
für Hydrobiologie, 4, 525-552.
Cranston, P.S., Fairweather, P. and Clarke, G. (1996) Biological indicators
of water quality, pp 143-154. In: Walker, J and Reuter, D.J. (eds) Indicators
of Catchment Health: a technical perspective. CSIRO, Melbourne.
Cummins, K.W. (1973). Trophic relations of aquatic insects. Ann. Rev.
Entomol. 18, 183-206.
Cummins, K.W. (1974). Structure and function of stream ecosystems,
BioScience, 24, 631-641.
Cummins, K.W. (1988). The study of stream ecosystems: a functional view,
p. 320-363. In: J. R. Pomeroy and J. J. Alberts (eds.). Concepts of
ecosystem ecology. Springer-Verlag, New York.
Cummins, K.W. and Klug, M.J. (1979). Feeding ecology of stream
invertebrates, Ann. Rev. Ecol. Syst. 10, 147-172.
Cummins, K.W., Coffman, W.P. and Roff, P.A. (1966) Trophic
relationships in a small woodland stream. Internationale vereinigung fur
theroetische und angewandte lomnologie. 16, 627-638.
Dadd, R.H. (1982) Insect nutrition: relevance to marine invertebrates. Pages
33-54 in G.D. Pruder, C.J. Langon and D.E Conklin (eds.) Proceedings of
the second International Conference on Aquaculture Nutrition. Division of
Continuing Education, Louisana State University, Baton Rouge.
Darley, W.M. (1982). Algal Biology: a Physiological Approach. Blackwell
Scientific Ppublications, Oxford.
Davis, W.S. and Simon, T.P. (1995) Introduction. Pages 3-6 in W.S. Davis
and T.P. Simon (eds). Biological assessment and criteria.Tools for water
resource planning and decision making. Lewis Publishers, Boca Raton,
Florida.
299
Davies, N.M., Norris, R.H. and Thoms, M.C. (2000) Prediction and
assessment of local stream habitat features using large-scale catchment
characteristics. Freshwater Biology, 45, 343-369.
Death, R.G. (2000) Invertebrate-substratum relationships. In Colier, K.J.
and Winterbourn, M.J. (eds.), New Zealand Stream Invertebrates: Ecology
and Implications for Management. New Zealand Limnological Society,
Christchurch: pp. 157-178.
Deegan, L., Peterson, B. and Portier, R. (1990) Stable isotopes and cellulose
activity as evidence for detritus as a food for juvenile gulf Menhaden.
Estuaries, 13, 14-19.
DELG 1998. S.I. No. 258/1998: Local Government (Water pollution) Act,
1977 (Water quality Standards For Phosphorus) Regulations, 1998.
Government Publications, Dublin.
Den Boer, P.J. (1968) Spreading of risk and stabilisation of animal numbers.
Acta Biotheoretica, 18, 165-194.
De Pauw, N. and Hawkes, H.A. (1993) Biological monitoring of river water
quality. In Walley, W.J. and S. Judd (eds), Proceedings of the Freshwater
Europe Symposium on River Water Quality Monitoring and Control.
Birmingham: 87-111.
De Pauw, N. and Vanhooren, G. 1983. Method for biological quality
assessment of watercourses in Belgium. Hydrobiologia,100,153-168.
DEV (Deutsches Institut fur Normung e.V.), 1987. Biologisch-okologische
Gewasserguteuntersuchung (Gruppe M): Allegemeine Hinweise, Planung
und Durchfuhrung von Fliebgewasserunter-suchungen (M1). In: Deutsche
Einheitsverfahren zur Wasser- Wasser-Abwasser- und
Schlammuntersuchng. VCH Verlagsgesellschaft mbH, Weinheim: 1-13.
DEV (Deutsches Institut fur Normung e.V.), 1992. Biologisch-okologische
Gewasserguteuntersuchung: Bestimmung des Saprobienindex (M2). In
Deutsche Einheitsverfahren zur Wasser-Abwasser- und
Schlammuntersuchng. VCH Verlagsgesellschaft mbH, Weinheim: 1-13.
Dietrich, M. and Anderson, N.H. (1995) Life cycles and food habits of
mayflies and stoneflies from temporary streams in western Oregon.
Freshwater Biology, 34, 47-60.
Dixit, S.S. and Smol, J.P., (1994) Diatoms as indicators in the
environmental monitoring and assessment program-surface waters (EMAP-
SW). Environmental Monitoring and Assessment, 31, 275-306.
Dodds, W.K., Smith, V.H. and Zander, B. (1997) Developing nutrient
targets to control benthic chlorophyll levels in streams: a case study of the
Clark Fork river. Water Research, 31, 1738-1750.
300
Downes, B.J., Lake, P.S. and Schreiber, E.S.G. (1995) Habitat structure and
invertebrate assemblages on stream stones: a multivariate view from the
riffle. Australian Journal of Ecology, 20, 502-514.
Dube, M.G., Culp, J.M. and Scrimgeour, G.J. (1997) Nutrient limitation and
herbivory: processes influenced by bleaching kraft pulp mill effluent.
Canadian Journal of Fisheries and Aquatic Sciences, 54, 2584-2595.
Edmunds, G.F., Jr., Jensen, S.L and Berner, L. (1976) The Mayflies of
North and Central America. University of Minnesota Press, Minneapolis.
Edwards, R.W. and Owens, M. (1965) The oxygen balance of streams,
ecology and the industrial society. Blackwell, Oxford. U.K.
Edwards, A.C. , Cook, Y., Smart, R. and Wade, A.J. (2000) Concentrations
of nitrogen and phosphorus in streams draining the mixed land-use Dee
Catchment, north-east Scothland. Journal Applied Ecology, 37, 159-170.
Egglishaw, H.J. and Mackay, D.W. (1967) A survey of the bottom fauna of
streams in the Scottish Highlands. Part 3. Seasonal changes in the fauna of
three streams. Hydrobiologia, 30, 305-334.
Elliott, J.M. (1967) The life histories and drifting of the Plecoptera and
Ephemeroptera in a Dartmoor stream. Journal of Animal Ecology, 36, 343-
362.
Elliott, J.M. (1977) Some methods for the statistical analysis of samples of
benthic invertebrates. Freshwater Biological Association, Windermere,
Special Publication 25.
Elliot, J.M. (1979) Some methods for the statistical Analysis of samples of
Benthic Invertebrates. Windemere: Freshwater Biological Association.
Elliott, J.M. and Humpesch, U.H. (1980) Eggs of Ephemeroptera. Annual
Report of the Freshwater Biological Association, 48, 41-52.
Elliott, J.M. and Humpesch, U.H. (1983) A key to the adults of the British
Ephemeroptera with notes on their ecology. Scientific Publications of the
Freshwater Biological Association, 47, 101.
Elliott, J.M., Humpesch, U.H. and Macan, T.T. (1988) Larvae of the British
Ephemeroptera: A key with ecological notes. Freshwater Biological
Association, Scientific Publication no 49.
Elwood, J.W., Newbold, J.D., Trimble, A.F. and Stark, R.W. (1981) The
limiting role of phosphorus in a woodland stream ecosystem: effects of
enrichment of leaf decomposition and primary producers. Ecology, 82,146-
158.
Engblom, E. (1996) Ephemeroptera, Mayflies. In Nilsson, A.N. (ed.),
Aquatic Insects of North Europe – A Taxonomic Handbook, Vol. 1. Apollo
Books, Stenstrup, 13-53.
301
Environmental Protection Agency (2002) In Environment in Focus. Key
Environmental Indicators for Ireland. Eds, Lehane, M., Le Bolloch, O. and
Crawley, P. Environmental Protection Agency, Wexford, Ireland.
Estep, M.L. and Vigg, S. (1985) Stable carbon and nitrogen isotope tracers
of trophic dynamics of natural populations and fisheries of the Lahotan lake
system. Can. J. Fish. Aquat. Sci., 42, 1712-1719.
European Comission (2000) Directive 2000/60/EC of the European
Parliament and of the Council – Establishing a framework for the
community action in the field of water policy. Brussels, Belgium, 23
October 2000.
Fahy, E. (1973) Observations on the growth of Ephemeroptera in fluctuating
and constant temperature conditions. Proc. R.I.A., 73, 133-149.
Feminella, J. W. and Hawkins, C.P (1995) Interactions between stream
herbivores and periphyton: a quantitative analysis of past experiments.
Journal of the North American Benthological Society, 14, 465-509.
Feminella, J.W. and Resh, V.H. (1991) Herbivorous caddisflies, macroalgae
and epilithic microalgae: dynamic interactions in a stream grazing system.
Oecological, 87, 247-256.
Finlay, K.J. (2001) Construction and evaluation of a new laboratory system
for rearing mayflies. In: Trends in Research in Ephemeroptera and
Plecoptera, pp. 241-250.
Fisher, S.G., Gray, L. J., Grimm, N.B. and Busch, D.E. (1982) Temporal
succession in a desert stream ecosystem following flash flooding. Ecol.
Monogr., 52, 93-110.
Flanagan, P.J. and Toner, P.F. (1972) The national survey of Irish rivers. A
report on water qulaity. Dublin. An Foras Forbartha.
Francoeur, S.N., Biggs, B.J.F, Smith, R.A. and Lowe, R.L. (1999) Nutrient
limitation of algal biomass accrual in streams. An in situ bioassay. Ecology,
66, 465-472.
Freeman, C.M., Lock, M.A., Marxsen, J. and Jones, S.E. (1990) Inhibitory
effects of high molecular weight dissolved organic matter upon metabolic
processes in biofilms from contrasting rivers and streams. Freshwater
Biology, 24, 159-166.
Frieberg, M. and Jacobsen, D. (1994) Feeding plasticity of two detritivore-
shredders. Freshwater Biology, 32, 133-142.
Frissel, C.A., Liss, W.J., Warren, C.E. and Hurley, M.D. (1986)
Hierarchical framework for stream habitat classification: viewing streams in
a watershed context. Environment Management, 10, 199-214.
302
Fuhs, W. (1969) Phosphorus content and the rate of growth in the diatoms
Cyclotella nana and Thalassiosira fluviatilis. J. Phycol. 5, 312-321.
Fuller, R.L., and Mackay, R. J. (1981) Effects of food quality on the growth
of three Hyrdopsyche species (Trichoptera: Hydropsychidae) Can. J. Zool.
59, 1133-1140.
Furse, M.T., Moss, D., Wright, J. F. and Armitage, P.D. (1984) The
influence of seasonal and taxonomic facts on the ordination and
classification of running water-sites in Great Britain and on the prediction of
their macro-invertebrate communities. Freshwater Biology, 14, 257-280.
Ghetti, P.F. (1997) Manuale di applicazione Indice Biotico Esteso (I.B.E.). I
macroinvertebrati nel controllo della qualita degli ambienti di acque
correnti. Provincia Autonoma di Trento, Agenzia provinciale per la
protezione dell’ambiente, 222pp.
Gifford, D.J. and Caron, D.A. (2000) Sampling, preservation, enumeration
and biomass of marine protozooplankton. In. Harris, R.P. et al. (eds) ICES
Zooplankton methodology manual. Academic Press, London, pp 193-221.
Giller, P.S. and Malmqvist, B. (1998) The biology of streams and rivers.
Oxford University Press.
Gordon ND., McMahon TA. and Finlayson BL. (1992) Stream hydrology.
An introduction for ecologists. John Wiley & Sons, Chichester. 526 pp.
Gore, J.A. (1985) The restoration of Rivers and Streams, 280 pp.
Butterworth, Stoneham.
Gray, L.J. and J.V. Ward. (1979) Food habits of stream benthos at sites of
differing food availability. American Midland Naturalist, 102, 157-167.
Green, R.H. (1979) Sampling design and statistical methods for
environmental biologists. John Wiley, New York, 257pp.
Gregory, S.V. (1983) Plant-herbivore interactions in stream ecosystems.
Pages 157-189. In Barnes J.R. and G.W. Minshall (eds.) Stream Ecology
Pleneum Press, New York: 157-189.
Grimm, N.B., and Fisher, S.G. (1986) Nitrogen limitation in a Sonoran
Desert stream. J. N. Am. Benthol. Soc., 5, 2-15.
Grimm, N.B. and Fisher, S.G. (1989) Stability of periphyton and
macroinvertebrates to disturbance by flash floods in a desert stream. Journal
of the North American Benthological Society, 8, 293-307.
Haefner, J.D. and Wallace, J.B. (1981) Shifts in aquatic insect populations
in a first-order southern Appalachian stream following a decade of old field
succession. Can. J. Fish. aquat. Sci. 38, 353-359.
303
Harker, J.E. (1952) A study of the life cycles and growth rates of four
species of mayflies. Proc. R. Ent. Soc. Lond. (A), 27, 77-85.
Hart, B.T., Maher, W. and Lawrence, I. (1999) New generation water
quality guidelines for ecosystem protection. Freshwater Biology, 41, 347-
359.
Harvey, C.J., Peterson, B.J., Bowen, W.B., Hershey, A.E., Miller, M.C.,
Deegan, L.A. and Finlay, J.C. (1998) Biological responses to fertilization of
Oksrukuyik Creek, a tundra stream. Journal of the North American
Benthological Society, 17, 190-209.
Hawkins, C.P. (1985) Food habits of species of Ephemerella mayflies
(Ephemeroptera, Insecta) in streams of Oregon. Am. Midl. Nat. 113, 343-
352.
Hawkins, C., Murphy, M and Anderson, N. (1982) Effects of canopy,
substrate composition and gradient on the structure of macroinvertebrate
communities in Cascade range streams of Oregon. Ecology, 63, 1840-1856.
Heathwaite, A.L., Johnes, P.J. and Peters, N.E. (1996) Trends in nutrients.
Hydrological Processes, 10, 263-293.
Hefti, D., Tomka, I. and Zurwerra, A. (1989) Revision of morphological and
biochemical characters of the European species of the Ecdyonurus
helveticus – group (Ephemeroptera, Heptageniidae). Bull. Soc. Entomol.
Sussie. 62, 329-344.
Hellawell, J.M. (1978) Biological Surveillance of Rivers: A Biological
Monitoring Handbook. Water Research Centre, Medmenham and
Stevenage.
Hellawell, J.M. (1986) Biological Indicators of Freshwater Pollution and
Environmental Management. Applied Science Publishers, London
Hering, D., Moog, O., Sandin, L. and Verdonschot, P.F.M. (2004) Overview
and application of the AQEM assessment system. Hydrobiologia, 516, 1-20.
Hildrew, A.G. (1992) Food webs and species interactions. The Rivers
Handbook: Hydrological and Ecological Principles, Vol. I (Eds P. Calow
and G.E. Petts), pp.309-329. Blackwell Scientific Publications, London.
Hill, W.R (1996) Effects of light. Algal Ecology: Freshwater Benthic
Ecosystems. Aquatic Ecology Series (eds. J.R. Stevenson, M.J. Bothwell and
R.L. Lowe), pp. 121-144. Academic Press, Inc., San Diego.
Hill, W.R. and Knight, A.W. (1987) Experimental analysis of the grazing
interaction between a mayfly and stream algae. Ecology, 68, 1955-1965.
Hill, W.R., Boston, H.L. and Steinman. (1992). Grazers and nutrients
simultaneously limit lotic primary productivity. Can. J. Fish. Aquat. Sci. 49,
504-512.
304
Hill, W.R., Ryon, M.G. and Schilling, E.M. (1995) Light limitation in a
stream ecosystem: responses by primary producers and consumers. Ecology,
76, 1297-1309.
Hilsenhoff, W.L. (1987) An improved biotic index of organic stream
pollution. Great Lakes Entomol. 20, 31-39.
Hobbie, J.E., Daley, R.J. and Jasper, S. (1977) Use of Nuclepore filters for
counting bacteria by fluorescence microscopy. Appl. Environ. Microbial.
33, 1225-1228.
Horkan, J.P.K. (1980) Interim report on eutrophication and related studies
of the Thurles area of the River Suir May-July 1978. Water Resources
Division, Dublin. 213pp
Horner, R.R. and Welch, E.B. (1981) Stream periphyton development in
relation to current velocity and nutrients. Can. J. Fish. Aquat. Sci. 38, 449-
457.
Horner, R.R., Welch, E.B. and Veenstra, R.B. (1983) Development of
nuisance periphytic algae in laboratory streams in relation to enrichment and
velocity. In Periphyton of Freshwater Ecosystems (Edited by Wetzel. R.G.)
pp.121-134. Junk, The Hague.
Hughes, R.M. (1995) Defining acceptable biological status by comparing
with reference conditions. Pages 31-47 in W.S. Davis and T.S. Simon
(editors)> Biological assessment and criteria. Lewis Publishers, Boca Raton,
Florida.
Humpesch, U.H. (1980) Effect of temperature on the hatching time of eggs
of five Ecdyonurus spp. (Ephemeroptera) from Austrian streams and
English streams, rivers and lakes. J. Anim. Ecol., 49, 317-333.
Humpesch, U.H. (1981) Effect of temperature on larval growth of
Ecdyonurus dispar (Ephemeroptera: Heptageniidae) from two English
lakes. Freshwater Biology, 11, 441-457.
Hunt, B.P. (1953) The life history and economic importance of a burrowing
mayfly, Hexagenia Limbata in southern Michigan lakes. Bulletin of the
Institute for Fisheries Research, University of Michigan, 4, 1-151.
Hunter, R.D. (1980) Effects of grazing on the quantity and quality of
freshwater aufwuchs. Hydrobiologia, 69, 251-259.
Hynes, H.B.N. (1956) Effect of drought on the fauna of a small mountain
stream in Wales. Verhandlungen der Internationalen Vereinigung fur
theoretische und angewandte Limnologie, 13, 826-833.
Hynes, H.B.N. (1960) The Biology of Polluted Waters. Liverpool Univ.
Press, Liverpool, England.
305
Hynes, H.B.N. (1961) The invertebrate fauna of a Welsh mountain stream.
Arch. Hydrobiol., 57, 344-388.
Hynes, H.B.N. (1966) The biology of polluted waters. Liverpool University
Press.
Hynes, H.B.N. (1968) Further studies on the invertebrate fauna of a Welsh
mountain stream. Archives of Hydrolobiology, 65, 360-379.
Hynes, H.B.N. (1970) The ecology of stream insects. Annual Review of
Entomology, 15, 25-41.
Hynes, H.B.N. (1970) The Ecology of Running Waters, University of
Toronto Press, Toronto.
Hynes, H.B.N. and Kaushik, N.K. (1969). The relationship between
dissolved nutrient salts and protein production in submerged autumnal
leaves. Verhandlungen Internationale Vereingung fur Theoretische und
Angewandte Limnologie, 17, 95-103.
Illies, J. (1978) Limnofauna Europaea. Gustav Fisher Verlag, Stuttgart, 2nd
Ed.
Inversen, T.M. (1974) Ingestion and growth in Sericostoma personatum
(Trichoptera) in relation to the nitrogen content of ingested leaves. Oikos,
25, 278-282.
Jeffers, J.N.R. (1998) Characterisation of river habitats and prediction of
habitat features using ordination techniques. Aquatic Conservation: Marine
and Freshwater Ecosystems, 8, 529-540.
Johnson, R.K. (1995) The indicator concept in freshwater biomonitoring.
Pages, 11-27, In: P.S. Cranston (ed.) Chironomids: from genes in
ecosystems. CSIRO, Melbourne.
Johnson, R.K. (1999) Regional representativeness of Swedish reference
lakes. Environmental Management, 23, 115-124.
Johnson, R.K. (2001) Defining reference condition and setting class
boundaries in ecological monitoring and assessment. Background document
for the EU-funded REFCOND project.
Jones, J.R., Smart, M.M. and Burroughs, J.N. (1984) Factors relating to
algal biomass in Missouri Ozark streams. Verhandlungen der internationale
Vereinigung fur theoretische und angewandte Limnologie, 22, 1867-1875.
Jowett, I.G., Richardson, J., Biggs, B.J.F., Hickey, C.W. and Quinn, J.M.
(1991) Microhabitat preferences of benthic invertebrates and the
development of generalised Deleatidium spp. habitat suitability curves,
applied to four New Zealand rivers. New Zealand J. mar. Freshwat. Res. 25,
187-199.
306
Kalff, J., Kling, H.J. Holmgren, S.H. and Welch, H.E. (1975)
Phytoplankton, phytoplankton growth and biomass cycles in an unpolluted
and in a polluted polar lake. Verh. Int. Ver. Limnol. 19, 487-495.
Karouna, N. and Fuller, R.L. (1992) Influence of four different grazers on
algal communities associated with clay tiles and leaf discs. Hydrobiologia,
245, 53-64.
Karr, J.R. (1991) Biological integrity: a long-neglected aspect of water
resource management. Ecological Applications, 1, 66-84.
Kelly, M.G. and Whitton, B.A. (1998). Biological monitoring of
eutrophication in rivers. Hydrobiologia, 384, 55-67.
Kelly-Quinn, M. and Bracken, J.J. (2000) The distribution of the
Ephemeroptera in Ireland. Occasional Publication of the Irish
Biogeographical Society 5.
Kelser, D.H. (1981) Periphyton grazing by Amnicola limnosa: an enclosure-
exclosure experiment. J. Freshwat. Biol. Ecol. 1, 51-59.
Kimmins, D.E. (1972) A revised key to the adults of the British species of
Ephemeroptera with notes on their ecology. Scient. Publs. Freshwat. Biol.
Assoc. no. 15, 75pp.
King, J.J. and Kelly, F.L. (1999) Experimental Drainage Maintenance
(EDM) and Urban Flood Relief Programmes (Impact of Works on fish stock
and habitat): Survey methodologies. Central Fisheries Board (unpublished
report).
Kjeldsen, K. (1996) Regulation of algal biomass in a small lowland stream:
field experiments on the role of invertebrate grazing, phosphorus and
irradiance. Freshwater Biology, 36, 535-546.
Kjeldsen, K., Iversen, T.M., Thorup, J. and Winding, T. (1998) Benthic
algal biomass in an unshaded first-order lowland stream: distribution and
regulation. Hydrobiologia, 377, 107-122.
Kling, G., Fry, B. and O’Brien, W.J. (1992) Stable isotopes and trophic
structure in arctic lakes. Ecology, 73, 561-566.
Klonowska, M. (1986) The food of some mayfly (Ephemeroptera) nymphs
from the streams of the Krakow-Czestochowa Jura (Southern Poland). Acta
Hydrobiol., 28, 181-197.
Knoben, R.A.E., Roos, C. and van Oirschot, M.C.M. (1995) Biological
assessment methods for watercourses. UN/ECE Task Force on Monitoring
and Assessment, Volume 3. RIZA report no. 95.066, Lelystad.
Kohmann, F., Binder, W. and Braun, P. (1993) Leitbilder fur die Erstellung
okologisch begrundeter Sanierungskonzepte kleiner Fliebgewasser,
Langfassung eines Vortrages, Tagung “Wasser” Berlin (30.4.1993).
307
Kondratieff, P.F and Simmons, Jr. G.M. (1985) Microbial colonisation of
seston and free bacteria in an impounded river. Hydrobiologia, 128, 127-
133.
Kostalos, M. and Seymour, R.L. (1976) Role of microbial enriched detritus
in the nutrition of Gammarus minus (Amphipoda). Oikos, 27, 512-516.
Krebs, C.J. (1989) Ecological methodology. Harper and Row, New York,
654pp.
Lake, P.S. and Barmuta, L.A. (1986) Stream benthic communities:
persistent presumptions and current speculations. Limnology in Australia.
(eds. P. DeDeckker and W.D. Williams), pp263-276. Commonwealth
Scientific and Industrial Research Organization, Melbourne.
Lamberti, G.A. and Moore, J.W. (1984) Aquatic insects as primary
consumers. The Ecology of Aquatic Insects (Eds V.H. Resh and D.M.
Rosenberg), pp.164-195, Praeger, New York.
Lamberti, G.A. and Resh, V.H. (1983) Stream periphyton and insect
herbivores: an experimental study of grazing by a caddisfly population.
Ecology, 64, 1124-1135.
Lamberti, G.A., Ashkenas, L.R., Gregory, S.V. and Steinman, A.D. (1987)
Effects of three herbivores on periphyton communities in laboratory
streams. Journal of North American Benthological Society, 6, 92-104.
Lamberti, G.A., Gregory, S.V., Ashkenas, L.R., Steinman, A.D., McIntire,
C.D. (1989) Productive capacity of periphyton as a determinant of plant
animal interactions in streams. Ecology, 70, 1840 1856.
Landa, V. (1962) Die Entwicklung der mitteleuropaischen Ephemeropteren.
International Congress of Entomology, 11, Vienna, 3, 250-254.
Landa, V. (1968) Developmental cycles of Central European
Ephemeroptera and their interrelations. Acta. entomologica. bohemoslovaca,
65, 276-284.
Ledger, M.E. and Hildrew, A.G. (1998) Temporal and spatial variation in
the epilithic biofilm of an acid stream. Freshwater Biology, 40, 665.
Ledger, M.E. and Hildrew, A.G. (2000) Herbivory in an acid stream.
Freshwater Biology, 43, 545-556.
Lembi, C.A., O’Neal, S.W. and Spencer, D.F. (1988) Algae as weeds:
economic impact, ecology, and management alternatives. pp. 455-481. In
Algae and Human Affairs. (Edited by Lembi, C.A. and Waaland, J.R.).
Cambridge, England).
Lenat, D.R. and Barbour, M.T. (1994) Using benthic macroinvertebrate
community structure for rapid, cost-effective, water quality monitoring:
308
rapid bioassessment. In Loeb, S.L. and Spacie, A. (eds), Biological
Monitoring of Aquatic Systems. Lewis Publishers, Boca Raton FL. 197-215.
Lillehammer, A., Brittain, J.E., Saltveit, S.J. and Nielsen, P.S. (1989) Egg
development, nymphal growth and life cycle, strategies in Plecoptera,
Holarctic Ecology, 12, 173-86.
Lloyd, D.S., Koenings, J.P. and LaPerriere, J.D. (1987) Effect of turbidity in
fresh waters of Alaska. North. Am. J. Fisheries. Mgmt., 7, 19-33.
Lock, M.A. (1992) Attached microbial communities in rivers. Aquatic
Microbiology: An Ecological Approach (ed. T. E. Ford), pp.113-139.
Blackwell Scientific Publications, Oxford.
Lock, M.A., Wallace, R.R., Costerton, J.W., Ventullo, R.M. and Charlton,
S.E. (1984) River epilithon: towards a structural-functional model. Oikos,
42, 10-22.
Lock, M.A., Ford, T.E., Hullar, M.A.J., Kaufman, M., Vestal, J.R., Volk,
G.S. and Ventullo, R.M. (1990) Phosphorus limitation in an arctic river
biofilm – A whole ecosystem experiment. Wat. Res., 24, 1545-1549.
Logan, P. (2001) Ecological quality assessment of rivers and integrated
catchment management in England and Wales. J.Limnol,. 60, 25-32.
Lohman, K., Jones, J.R. and Baysinger-Daniel, C. (1991) Experimental
evidence for nitrogen limitation in a northern Ozark stream. J. N. Am.
Benthol. Soc. 19, 14-23.
Lohman, K., Jones, J.R. and Perkins, B.D. (1992) Effects of nutrient
enrichment and food frequency on periphyton biomass in northern Ozark
streams. Canadian Journal of Fisheries and Aquatic Sciences, 49, 1198-
1205.
Lopez, G.R. and Kofoed, L.H. (1980) Episammic browsing and deposit-
feeding in mud snails (Hydrobiidae). J.Mar.Sci. 38, 585-597.
Lowe, R.L. and Hunter, R.D. (1988) Effects of grazing by Physa integra on
periphyton community structure. J. North. Am. Benthol. Soc. 7, 29-36.
Lucey, J., Bowman, J.J., Clabby, K.J., Cunningham, P., Leane, M.,
MacCarthaigh, M., McGarrigle, M.L., Toner, P.F. (1999) Water quality in
Ireland 1995-1997. Environmental Protection Agency, Wexford.
Ludwig, J.A. and Reynolds, J.F. (1988) Statistical Ecology. John Wiley,
New York, 337pp.
Lugthart, G.J. and Wallace, J.B. (1992) Effects of disturbance on benthic
functional structure and production in mountain streams. Journal of the
North American Benthological Society, 11, 138-164.
309
Lund, J.W.G. (1969) Phytoplankton, p. 306-330. In Eutrophication: Causes,
consequences, correctives. Natl. Acad. Sci. Washington, D.C.
Lydy, M.J., Crawford, C.G. and Frey, J.W. (2000) A comparison of selected
diversity, similarity and biotic indices for detecting changes in benthic-
invertebrate community structure and stream quality. Arch. Environ.
Contam. Toxicol., 39, 469-479.
Macan, T.T. (1957) The life histories and migrations of the Ephemeroptera
in a stony stream. Transactions of the Society of British Entomologists, 12,
129-156.
Macan, T.T. (1958) Methods of sampling the bottom fauna in stony streams.
Communications. International Association of Theoretical and Applied
Limnology, 8, 1-21.
Macan, T.T. (1970) Biological Studies of the English Lakes. Longman
Group Limited, London.
Macan, T.T. (1979) A key to the nymphs of the British species of
Ephemeroptera with notes on their ecology. Freshwater Biological
Association Scientific Publication 20, Ambleside, UK.
Macan, T.T. (1981) Life histories of some species of Ecdyonurus
(Ephemeroptera) in the River Lune, North-Western England.
Macan, T.T. and Maudsley, R. (1968) The insects of the stony substratum of
Windermere. Transactions of the Society for British Entomology, 18, 1-18.
Maciolek, J. A., and Tunzi, M.G. (1968). Microseston dynamics in a simple
Sierra Nevada lake-stream system. Ecology, 49, 60-75.
MacKenthun, K.M. (1968) The phosphorus problem. J. Am. Water. Works.
Assoc. 60, 1047-1053.
Madden, C.P., Austin, A.D. and Suter, P.J. (1995) Pollution monitoring
using chironomid larvae: what is a deformity? Pages 89-100. In P.S.
Cranston (ed.) Chironomids: from genes to ecosystems. CSIRO, Melbourne.
Maddock, I. (1999) The importance of physical habitat assessment for
evaluating river health. Freshwater Biology, 41, 373-391.
Madsen, B.L. (1972). Detritus on stones in small streams. Memorie
Dell’instituto Italiano Di Idrobiologia, 29, 385-403.
Maher W., Batley, G.E. and Lawrence I. (1999) Assessing the health of
sediment ecosystems: Use of chemical measurements. Freshwater Biology,
41, 361-372.
Manuel, C.Y. and Minshall, G.W. (1978) Limitations on the use of
microcosms for predicting algal response to nutrient enrichment in lotic
310
systems. pp 645-667.Department of Biology, Idaho State University,
Pocatello, Idaho.
Margalef, R. (1958) Information theory in ecology. Entomol. Systematics, 3,
36-71.
Marten, M. and Reusch, H. (1992) Anmerkungen zum DIN
‘Saprobienindex’ (38 410 Teil 2) und Forderung alternativer Verfahren.
Natur Landschaft 67, 544-547.
Martinez, N.D. (1993) Effects of resolution on food web structure. Oikos,
66, 403-412.
Mason, C.F. (1991) Biology of Freshwater Polluiton. Longman Scientific
and Technical, Essex.
Maurice, C.G., Lowe, R.L., Burton, T.M. and Stanford, R.M. (1987)
Biomass and compositional changes in the periphytic community of an
artificial stream in response to lowered pH. Water, Air and Soil Pollution.
33, 165-177.
McAuliffe, J.R. (1984) Resource depression by a stream herbivore: effects
of distribution and abundance of other grazers. Oikos, 42, 327-346.
McCormick, P.V. and Stevenson, R.J. (1991) Grazer control of nutrient
availability in the periphyton. Oecologia, 86, 287-291.
McGarrigle, M.L. (1983) Development of Water Quality Management
Methods Based on systems Analysis (River Suir). EEC Contract 6612/4,
Department of the Environment and An Foras Forbartha, Dublin. 42pp.
McGarrigle, M.L. (1984a) Development of a water quality model for Irish
Rivers. Preliminary studies on the River Suir. WR/R16, Water Resource
Division, An Foras Forbartha, Dublin.
McGarrigle, M.L. (1984b) Experiments with the River Model. Ir. J.
Environ. Sci. 3, 40-47.
McGarrigle, M.L. (1998) Impact of eutrophication on Irish river water
quality. In: Eutrophication in Irish Waters (Ed. J.G. Wilson) pp 1-9 Royal
Irish Academy.
McGarrigle, M.L. (2001) In Went Memorial Lecture 2001: Eutrophication
of Irish Waters – From Science to legislation and management.
McGarrigle, M.L. and Lucey, J. (1989) Waste assimilative capacity survey
of the Athboy River. Report to Meath County Council. Environmental
Research Unit.
McGarrigle, M.L. and Lucey, J. (1990) Assimilative capacity survey of the
Athboy River. Report to Meath County Council. Environmental Research
Unit.
311
McGarrigle, M.L., Horkan, J.P.K. and Lucey, J. (1987) Water quality
survey of the Magherarney river. Report to Monaghan County Council,
WR/C132, Water Resources Division, An Foras Forbartha, Dublin.
McGarrigle, M.L., O’Grady, M., Champ, W.S.T. and McCarthy, T.K.M.
(1998) The Moy Catchment pp212-236. In Moriarty, C. (Ed.) Studies of
Irish rivers and Lakes. Marine Institute, Dublin. 279pp. ISBN0529 191 29.
McGarrigle, M.L., Bowman, J.J., Clabby, K.J., Lucey, J., Cunningham, P.,
MacCarthaigh, M., Keegan, M. Cantrell, B., Lehane, M., Clenaghan, C. and
Toner, P.F. (2002) Water Quality in Ireland 1998-2000. Envionmental
Protection Agency.
Merritt, R.W. and Cummins, K.W. (1984) An introduction to the aquatic
insects of North America, Second Edition. Kendall/Hunt Publishing
Company.
Merritt, R.W., Cummins, K.W. and Resh, V.H. (1996) Design of Aquatic
Insect Studies: collecting, sampling and rearing procedures, pp. 12 –28. In:
R.W. Merritt and K.W. Cummins (eds.) An introduction to the Aquatic
Insects of North America. 3rd
edn. Kendall Hunt Publishing Co.Iowa.
Metcalfe, J.L. (1989) Biological water quality assessment of running water
based on macoinvertebrate communities: history and present status in
Europe. Environ. Pollut., 60, 101-139.
Meulemans, J.T. (1987) A method for measuring selective light attenuation
within a periphytic community. Arch. Hydrobiol. 109, 139-145.
Meyer, J.L., McDowell, W.H., Bott, T.L., Elwood, J.W., Ishizaki, C.,
Melack, J.M. Peckarsky, B.L., Peterson, B.J. and Rublee, P.A. (1988)
Elemental dynamics in streams. Journal of the North American
Benthological Society, 7, 410-432.
Mihuc, T.B. (1997) The functional trophic role of lotic primary consumers:
generalist versus specialists strategies. Freshwater Biology, 37, 455-462.
Mihuc, T.B. and Minshall, G.W. (1995) Trophic generalists vs. trophic
specialists: implications for food web dynamics in post-fire streams.
Ecology, 76, 2361-2372.
Mihuc, T.B. and Toetz, D. (1994) Determination of diets of alpine aquatic
insects using stable isotopes and gut analysis. American Midland Naturalist,
131, 146-155.
Minshall, G.W. (1967) Role of allochthonous detritus in the trophic
structure of a woodland springbrook community. Ecology, 48, 139-149.
Minshall, G.W. (1984) Aquatic insect-substratum relationships. In: The
Ecology of Aquatic Insects (eds. V.H. Resh and D.M. Rosenberg), pp. 258-
400. Praeger Publishers, pp, New York.
312
Momot, W.T. and Jones, P.D. (1977) The relationship between biomass,
growth rate and annual production in the crayfish. Oronectes virilis
Freshwater Crayfish, 3, 3-31. (3rd
International Symposium on Freshwater
Crayfish, Finland)
Moog, O. (1993) Makrozoobenthos als Indikator bei okologischen
Fragestellungen, Landschaftswasserbau 15, 103-143, Wien.
Moog, O. (1994) Okologische Funktionsfahigkeit des aquatiscehn
Lebensraumes, Wiener Mitt. 120, 15-59.
Moog, O. (1995) Fauna Aquatica Austriaca. Version 1995.
Wasserwirtschafts-kataster, Bundesministerium fur Land-und
Forstwirtschaft, Wien. A comprehensive species inventory of Austrian
aquatic organisms with ecological notes.
Moog, O., Bauernfeind, E. and Weichselbaumer, P. (1997) The use of
Ephemeroptera as saprobic indicators in Austria. In: Landolt, P. and M,
Sartori (eds), Ephemeroptera and Plecoptera: Biology-Ecology-Systematics
(= Proc. 8th
Int. Conf. Ephemeroptera, Lausanne, 1995). Fribourg: 254-260.
Moriarty, C. (1990) Fish Kills in Ireland in 1990. Fishery Leaflet 149.
Dublin, Department of the Marine, Roinn na Mara.
Moss, B., Stephen, D., Alvarez, C., Becares, E., Van de Bund, W., Van
Donk, E., De Eyto, E., Feldmann, T., Fernandez-Alaez, C., Fernandez-
Alaez, M., Franken (2001) The determination of ecological quality in
shallow lakes – a tested system (ECOFRAME) for implementation of the
European Water Framework Directive.1-55.
Muhar, S. and Jungwirth, M. (1998) Habitat integrity of running waters –
assessment criteria and their biological relevance. Hydrobiologia, 386, 195-
202.
Mullholland, P.J., Steinman, A.D., Palumbo, A.V., Elwood, J.W. and
Kirschtel, D.B. (1991) Role of nutrient cycling and herbivory in regulating
periphyton communities in laboratory streams. Ecology, 72, 966-982.
Mundie, J.H., Simpson, K.S. and Perrin, C.J. (1991) Responses of stream
periphyton and benthic insects to increases in dissolved inorganic
phosphorus in a mesocosm. Canadian Journal of Fisheries and Aquatic
Sciences, 48, 2061-2072.
Newcombe, C.P. and MacDonald, D.D (1991) Effects of suspended
sediments on aquatic ecosystems. North. Am. J. Fisheries. Mgmt., 11, 72-82.
Newell, R.L. and Minshall, G.W. (1978) Like history of a multivoltine
mayfly. Tricorythodes minutus: an example of the effect of temperature on
the life cycle. Annals of the Entomological Society of America, 71, 876-881.
313
Ní Chatháin, B., (2002) Investigation of the ecology of benthic algae in the
river Deel, South-West Ireland, Ph.D. thesis, University of Limerick,
Ireland, 273 pp.
Ní Chatháin B., Harrington T. J., McCarthy J., Kelly-Quinn, M., Bradley,
C., and Baars, J.R. (2004) Preliminary results from the investigation of
benthic diatoms from potential reference river sites in Ireland. The
International Journal of Oceanological and Hydrobiological Studies. In
press.
Nixon, S.C., Mainstone, C.P., Milne, I., Inversen, T.M., Kristensen, P.,
Jeppesen, E. Friberg, N., Jensen, A. and Pedersen, F. (1996) The
harmonised monitoring and classification of ecological quality of surface
waters in the European Union. Draft Final Report. No. CO., 4096.
Norris, R.H. and Norris, K.R. (1995) The need for biological assessment of
water quality: Australian perspective. Aust. J. Ecol., 20, 1-6.
Norris, R.H. and Thoms, M.C. (1999) What is river health? Freshwater
Biology, 41, 197-209.
Norris, R.H., McElravy, E.P. and Resh, V.H. (1992) The sampling problem.
The Rivers Handbook, Hydrological and Ecological Principles (eds P.
Callow and G.E. Petts), pp.282-306. Blackwell Scientific Publications,
Oxford.
OECD (Organisation for the Economic Cooperation and Development)
(1982) Eutrophication of waters. Monitoring assessment and control. Final
report. OECD Coopertive Programme on monitoring of Inland Waters
(Eutrophication Control), Environmental Directorate, OECD, Paris, 154 p.
Odum, H.T. (1956) Primary production in flowing waters. Limnol.
Oceanogr., 1, 102-117.
Ofenbock, T., Moog, O., Gerritsen, J. and Barbour, M. (2004) A stressor
specific multimetric approach for monitoring running waters in Austria
using benthic macroinvertebrates. Hydrobiologia, 516, 251-268.
Otto, C. (1974) Growth and energetics in a larval population of
Potamophylax cingulatus (Steph.) (Trichoptera) in a South Swedish stream.
J. Animl. Ecol. 43, 339-361.
Pahl-Wostl, C. (1993) Food webs and ecological networks across temporal
and spatial scales. Oikos, 66, 415-432.
Paine, R.T. (1988) Food webs: road maps of interactions or grist for
theoretical development. Ecology, 69, 1648-1654.
Pan, Y. and Lowe, R.L. (1994) Independent and interactive effects of
nutrients and grazers on benthic algal community structure. Hydrobiologia,
291, 201-209.
314
Pan, Y., Stevenson, R.J., Hill, B.H., Herlihy, A.T. and Collins, G.B. (1996)
Using diatoms as indicators of ecological conditions in lotic systems: a
regional assessment. J. N. Am. Benth. Soc., 15, 481-495.
Pantle, R. and Buck, H. (1955) Die biologische Uberwachung der Gewasser
und die Darstellung der Ergebnisse. Bes. Mitt. Dt. Gewasserkundl. Jb. 12,
135-143.
Parsons, M., and Norris, R.H., (1996) The effects of habitat-specific
sampling on biological assessment of water quality using a predictive
model. Freshwater Biology. 36, 419-434.
Patrick, R. (1948) Factors affecting the distribution of diatoms. Botanical
Review, 14, 473-524.
Paul, B.J. and Duthie, H.C. (1989) Nutrient cycling in the epilithon of
running waters. Can. J. Bot. 67, 2302-2309.
Peeters, E.T.H.M., Gardeniers, J.J.P. and Tolkamp, H.T. (1994) New
methods to assess the ecological status of surface waters in the Netherlands.
Part 1: Running waters. Verh. Int. Ver. Limnol., 1914-1916.
Pentecost, A. (1984) Introduction to Freshwater Algae. Richmond.
Richmond Publishing Co.
Perrin, C.J., Bothwell, M.L. and Slaney, P.A. (1987) Experimental
enrichment of a coastal stream in British Columbia: effects of organic and
inorganic addition on atrophic periphyton production. Can. J. Fish. Aquat.
Sci. 44, 1247-1256.
Peter, R.E. (1979) The brain and feeding behaviour. In: Fish Physiology
(Eds. W. S. Hoar, D.J. Randall and J.R. Brett), Vol. 8. pp.121-159.
Academic Press, London.
Peterson, C. G. and Stevenson, R. J. (1992) Resistance and resilience of
lotic algal communities: importance of disturbance timing and current.
Ecology, 73, 1445–1461.
Peterson, B.J., Bahr, M. and Kling, G.W. (1997) A tracer investigation of
nitrogen cycling in a pristine tundra river. Canadian Journal of Fisheries
and Aquatic Sciences, 54, 2361-2367.
Peterson, B.J., Hobbie, J.E., Hershey, A.E., Loch, M.A., Ford, T.E., Vestal,
J.R., McKinley, V.L., Hullar, M.A.J., Miller, M.C., Ventullo, R.M. and
Volk, G.S. (1985) Transformation of a tundra river from heterotrophy to
autotrophy by addition of phosphorus. Science, 229, 1383-1386.
Plafkin, J.L., Barbour, M.T., Porter, K.D., Gross, S.K. and Hughes, R.M.
(1989) Rapid Bioassessment Protocols for Use in Streams and Rivers:
Benthic Macroinvertebrates & Fish. US Environmental Protection Agency
Assessment and Watershed Protection Division, Washington, D.C.
EPA/440/4-89/001.
315
Polis, G.A. (1994). Food webs, trophic cascades and community structure.
Australian Journal of Ecology, 19, 121-136.
Porter, K.G. and Feig, Y.S. (1980) The use of DAPI for identifying and
counting aquatic microflora. Limnol. Oceanogr. 25, 943-948.
Power, M.E., Stout, R.J., Cushing, C.E., Harper, P.P., Hauer, F.R.,
Matthews, W.J., Moyle, P.B., Statzner, B and Wais De Badgen, I.R. (1988).
Biotic and abiotic controls in river and stream communities. J.N. Am.
Benthol. Soc., 7, 4, 456-479.
Pringle, C.M. and Bowers, J. (1984) An in situ substratum fertilization
technique: Diatom colonization on nutrient-enriched sand substrata.
Canadian Journal of Fisheries and Aquatic Sciences, 41, 1247-1251.
Quinn, J.M. and Hickey, C.W. (1990) Magnitude of effects of substrate
particle size, recent flooding and catchment development on benthic
invertebrates in 88 New Zealand rivers. New Zealand J. Mar. Freshwat.
Res. 24, 411-427.
Rapport, D.J. (1991) Myths in the foundations of economics and ecology.
Biological journal of the Linnaean Society, 44, 185-202.
Rassoulzadegan, F. and Sheldon, R.W. (1986) Predator-prey interactions of
nanozooplankton and bacteria in an oligotrophic marine environment.
Limnol. Oceanogr. 31, 1010-1021.
Rau, G.H. (1980) Carbon-13/Carbon-12 variation in subalpine lake aquatic
insects: food source implications. Can. J. Fish. Aquatic. Sci., 37, 742-746.
Raven, P. J., Boon, P. J., Dawson, F. H., and Ferguson, A. J. D. (1998)
Towards an integrated approach to classifying and evaluating rivers in the
UK. Aquatic Conservation-Marine and Freshwater Ecosystems, 8, 383-393.
Raven, P.J., Fox, P.J.A., Everard, M., Holmes, N.T.H. and Dawson, F.H.
(1997) ‘River Habitat Survey: a new system for classifying rivers according
to their habitat quality’, in Boon, PJ and Howell, D.L.(Eds), Freshwater
Quality: Defining the indefinable?, The Stationery Office, Edinburgh, 215 –
234.
Rawlinson, D. (1939) Studies on the life-history and breeding of
Ecdyonurus venosus (Ephemeroptera). Proceedings of the Zoological
Society of London, Series B. 109, 377-450.
Reckhow, K.H. and Chapra, S.C. (1983) Engineering Approaches for Lake
Management, Vol.1, p.340. Ann Arbor Science, Butterworth Publishing
Co., Woburn, MASS.
Redfield, A.C. (1958) The boloigical control of chemical factors in the
environment. Am. Sci. 46, 205-221.
316
REFCOND (2003) Final version 7.0. Guidance on establishing reference
conditions and ecological class status boundaries for inland surface waters.
CIS Working group 2.3. p 1- 93.
Resh, V.H. and Unzicker, J.D. (1975) Water quality monitoring and aquatic
organisms: the importance of species identification. J. Wat. Pollut. Cont.
Fed., 47, 9-19.
Resh, V.H., Norris, R.H. and Barbour, M.T. (1995) Design and
implementation of rapid assessment approaches for water resource
monitoring using benthic macroinvertebrates. Austral. J. Ecol., 20, 108-121.
Reusch, H. (1985) Limnofaunistische Untersuchungen uber die Eintags-,
Stein-und Kocherfliegen (Ephemeroptera, Plecoptera, Trichoptera) der
Ortze (Niedersachsisches Tiefland). Jb. Naturw. Verein Fstm. Lutenburg,
37, 117-139.
Reusch, H. (1995) Planungsrelevante Aufbereitung und Bewertung
faunistisch-okolgischer Daten vom Makrozoobenthon in Fliebgewassern.
Sch.-R. Landschaftspfl. Natursch, 43, 31-42.
Reynoldson, T.B., Bailey, R.C., Day, K.E. and Norris, R.H. (1995)
Biological guidelines for the freshwater sediment based on Benthic
Assessment of SedimenT (the BEAST) using a multivariate approach for
predicting biological state. Australian Journal of Ecology, 20, 198-219.
Reynoldson, T.B., Norris, R.H., Resh, V.H., Day, K.E. and Rosenberg,
D.M. (1997) The reference condition: a comparison of multimetric and
multivariate approaches to assess water-quality impairment using benthic
macroinvertebrates. Journal of the North American Benthological Society,
16, 833-852.
Rhee, G.Y. (1973) A continuous culture study of phosphate uptake, growth
rate and polyphosphate in Scenedesmus sp. J. Phycol. 9, 495-506.
Rhee, G Y. (1978) Effects of N:P atomic ratios and nitrate limitation on
algal growth, cell composition and nitrate uptake. Limnology and
Oceanography, 23, 10-25.
Richards, C. and Minshall, G.W. (1988) The influence of periphyton
abundance on Baetis bicaudatus distribution and colonisation in a small
stream. Journal of the North American Benthological Society, 7, 77-86.
Richards, C., Host, G.E. and Arthur, J.W. (1993) Identification of
predominant environmental factors structuring stream macroinvertebrate
communities within a large agricultural catchment. Freshwater Biology, 29,
285-294.
Richards, C., Johnson, L.B. and Host, G.E. (1996) Landscape-scale
influences on stream habitats and biota. Canadian Journal of Fisheries and
Aquatic Sciences, 53, 295-311.
317
Ricker, W.E. (1979) Growth rates and models. In: Fish Physiology (Eds. W.
S. Hoar, D.J. Randall and J.R. Brett), Vol.8 pp. 677-743. Academic Press,
London.
Roche, W. (1996) River Crana Habitat Monitoring Programme: Substrate
Analysis Study No.1- 1994 Results. Central Fisheries Board, Dublin.
Rosenberg, D.M. and Resh, V.H. (1993) Introduction to Freshwater
Biomonitoring and Benthic Macroinvertebrates. In Freshwater
Biomonitoring and Benthic Macroinvertebrates (editors D.M. Rosenberg
and V.H. Resh). Chapman and Hall, New York.
Rosenberg, D.M. Danks, H.V. and Lehmkuhl, D.M. (1986) Importance of
insects in environmental impact assessment. Envir. Manage. 10, 773-783.
Rotschein, J. (1982) Nove metody stanoveni indexu saprobity (New
methods of saprobic index assessment) Vodni hospodarstvi B, 6, 159-162.
Rott, E., Duthie, H.C., and Pipp, E., (1998) Monitoring organic pollution
and eutrophication in the Grand River, Ontario, by means of diatoms. Can.
J. Fish. Aquat. Sci., 55, 1443-1453.
Rounick, J.S. and Winterbourn, M.J. (1983) Formation structure and
utilisation of stone surface organic layers in two New Zealand streams.
Freshwater Biology, 13, 57-72.
Rounick, J.S., Winterbourn, M.J. and Lyon, G.L. (1982) Differential
utilization of allochthonous and autochthonous inputs by aquatic
invertebrates in some New Zealand streams: a stable carbon isotope study.
Oikos, 39, 191-198.
Ryan, P.A. (1991) Environmental effects of sediment on New Zealand
streams: a review. New Zealand J. mar. Freshwat. Res. 25, 207-221.
Ryder, G.I. (1989) Experimental studies of the effects of fine sediments on
lotic invertebrates. Ph.D. thesis, University of Otago, Dunedin, New
Zealand: 216 pp.
Ryding, S.O. and Rast, W. (1989) The Control of Eutrophication of Lakes
and Reservoirs. Vol. 1, p. 314. UNESCO, Paris and The Parthenon
Publishing Group, United Kingdom.
Ryther, J. and Dunstan, W.M. (1971) Nitrogen, phosphorus and
eutrophication in the coastal and marine environment. Science, 171, 1008-
1013.
Sabater, S., Sabater, F. and Armengol, J. (1988) Relationships between
diatom assemblages and physico-chemical variables in the River Ter (NE
Spain), Int. Revue gesamt. Hydrobiology, 73, 171-179.
Sangpradub, N., Giller, P. and O’Connor, J (1999) Life history patterns of
stream-dwelling caddis. Arch. Hydrobiol. 146, 471-493.
318
Sas, H. (1989) Lake Restoration by reduction of Nutrient Loading:
Expectations, Experiences, Extrapolations. Academia Verlag, Richarz.
Scarsbrook, M.R. and Townsend, C.R. (1993) Stream community structure
in relation to spatial and temporal variation: A habitat templet study of two
contrasting New Zealand streams. Freshwat. Biol., 29, 395-410.
Schindler, D.W. (1975) Whole-lake eutrophication experiments with
phosphorus, nitrogen and carbon. Verh. Int. Ver. Limnol. 19, 3221-3231.
Schindler, D.W. (1985) The coupling of elemental cycles by organisms:
evidence from whole-lake chemical perturbations. In Chemical Processes in
Lakes (Edited by Stumm W.). Wiley-Interscience, New York.
Schoener, T.W. (1974) Resource partitioning in ecological communities.
Science, 185, 27-38.
Schoenly, K., Beaver, R.A. and Heumier, T.A. (1991). On the trophic
relations of insects: a food-web based approach. American Naturalist, 137,
597-638.
Schweder, H. (1992) Neue Indices fur die Bewertung des okologischen
Zustandes von Fliebgewassern, abgeleitet aus der Makroinvertebraten-
Ernahrungstypologie, Limnologie aktuell Bd. 3. 353-377, G. Fischer Vlg.
Stuttgart-New York.
Schwenneker, B.W. and Hellenthal, R.A. (1984) Sampling considerations in
using stream insects for monitoring water quality. Environmental
Entomology, 13, 741-750.
Scrimgeour, G.J., Culp, J.M., Bothwell, M.L., Wrona, F.L. and McKee,
M.H. (1991) Mechanisms of algal patch depletion: importance of
consumptive and non-consumptive losses in mayfly diatom systems.
Oecologia, 85, 343-348.
Shapas, T.J. and Hilsenhoff, W.L. (1976) Feeding habits of Wisconsin’s
predominant lotic Plecoptera, Ephemeroptera and Trichoptera. Great Lakes
Entomologist, 9, 175-178.
Sharpley, A., Foy, B., and Withers, P. (2000) Practical and innovative
measures for the control of agricultural phosphorus losses to water: An
overview. J. Environ. Qual, 29, 1-9.
Simpson, J. and Norris, R.H. (2000) Biological assessment of water quality:
development of AUSRIVAS models and outputs. In: Assessing the
Biological Quality of Freshwaters: RIVPAC and Similar Techniques (eds
J.F.Wright, Sutcliffe, D.W. and Furse, M.T., PP. 125-142. Freshwater
Biological Association and Environment Agency, U.K.
Skriver, J., Friberg, N. and Kirkegaard, J. (2001) Biological assessment of
running waters in Denmark: introduction of the Danish Stream Fauna Index
319
(DSFI). Verhandlungen der Internationale Vereinigung für Theoretische
und Angewandte Limnologie, 27, 1822-1830.
Sladecek, V. (1973) System of water quality from the biological point of
view. Arch. Hydrobiol. Beih. Ergenbnisse. Limnol., 7, 1-218.
Sladecek, V., Hawkes, H.A., Alabaster, J.S., Daubner, I., Notlich, I., Solbe,
J.F.D. and Uhlamann, D. (1982). Biological examination. In Examination of
Water for Pollution Control. A reference handbook. Vol. 3. Biological,
Bacteriological and Virological Examination. (editors M.J Suess).
Pergamon Press, Oxford.
Smith, V.H. (1979) Nutrient dependence of primary productivity in lakes.
Limnology and Oceanography, 24, 1051-1064.
Smith, V.H. (1982) The nitrogen and phosphorus dependence of algal
biomass in lakes: an empirical and theoretical analysis. Limnol. Oceanogr.
27, 1101-1112.
Smock, L.A. (1996) Macroinvertebrate movements: drift, colonisation and
emergence, pp. 371-390. In: F.R.Hauer and G.A. Lamberti (eds.). Methods
in Stream Ecology. Academic Press, San Diego.
Smock, L.A. and Roeding, C.E. (1986) The trophic basis of the
macroinvertebrate community of a southeastern U.S.A. blackwater stream.
Holartic Ecology, 9, 165-174.
Southwood, T.R.E. (1977) Habitat, the templet for ecological strategies?
Journal of Animal Ecology, 46, 337-365.
Southwood, T.R.E. (1988) Tactics, strategies and templets. Oikos, 52, 3-18.
Sowa, R. (1975) Ecology and biogeography of mayflies (Ephemeroptera) of
running waters in the polish part of the carpathians. 2. Life cycles. Acta.
Hydrobiol., 17, 319-359.
Sowa, R. (1979) Le developpement des Ephemeropteres de la riviere
Dunajec aux environs de Piening. Proceedings of the 2nd
International
Conference on Ephemeroptera, 125-131.
Standard Methods (2001) For the examination of water and wastewater
edition. 17th
edition.
Standard Methods (2002) For the examination of Water and Wastewater,
20th
edition.
Stanford, J.A. and Ward, J.V. (1983) Insect species diversity as a function
of environmental variability and disturbance in stream systems. Stream
Ecology, Application and Testing of General Ecological Theory (eds J.R.
Barnes and G.W. Minshall), pp. 265-278. Plenum Press, New York.
320
Steinman, A.D. (1996) Effects of grazers on freshwater benthic algae. Algal
Ecology Series (eds J. R. Stevenson, M. I. Bothwell and R.L. Lowe),
pp.341-366. Academic Press, Inc., San Diego.
Steinman, A.D., Mulholland, P.J. and Kirschtel, D.B. (1991) Interactive
effects of nutrient reduction and herbivory on biomass, taxonomic structure
and P uptake in lotic periphyton communities. Can. J. Fish. Aquat. Sci., 48,
1951-1959.
Steinman, A.D., McIntire, C.D., Gregory, S.D. and Lamberti, G.A (1989)
Effects of irradiance and grazing on lotic algal assemblages. Journal of
Phycology, 25, 478-485.
Steinman, A.D., McIntire, C.D., Gregory, S.V. and Lamberti, G.A. and
Ashkenas, L.R. (1987) Effects of herbivore type and density on taxonomic
structure and physiognomy of algal assemblages in lab streams. J. North.
Am. Benthol. Soc. 6, 175-188.
Stevenson, R.J. (1996) An introduction to algal ecology in freshwater
benthic habitats. Algal Ecology: Freshwater Benthic Ecosystems, Aquatic
Ecology Series (eds. J. R. Stevenson. M. I. Bothwell and R. L. Lowe), pp.3-
30. Academic Press Inc., San Diego, CA.
Stockner, J.G. and Evans, D. (1974) Field and laboratory studies on the
effects of nitrogen, phosphorus and NTA on attached algal communities.
Fish. Mar. Serv. Res. Dev. Tech. Rep., 416, 109p.
Stockner, J.G. and Shortreed, K.R.S. (1976) Autotrophic production in
Carnation Creek, a coastal rainforest stream on Vancouver island, British
Columbia. J. Fish. Res. Board Can. 33, 1553-1563.
Stockner, J.G. and Shortreed, K.R.S. (1978) Enhancement of autotrophic
production by nutrient addition in a coastal stream on Vancouver Island. J.
Fish. Res. Board Can., 35, 28-34.
Stout, B.M., Benfield, E.M. and Webster, J.R. (1993) Effects of forest
disturbance on shredder production in southern Appalachian streams.
Freshwater Biology, 29, 59-69.
Struder-Kypke et al. in prep In Guide to UK Coastal Planktonic Ciliates
(2001) DJS Montagnes, University of Liverpool.
Studemann, D. Landolt, P., Sartori, M., Hefti, D. and Tomka, I. (1992) Soc.
Entomol Suisse (ed.), Ephemeroptera. Insecta helvetica, Fauna 9. Fribourg,
1-174.
Swanson, F.J. (1980) Geomorphology and ecosystems. In: Forests: Fresh
Perspectives from Ecosystem Analysis. Proceedings of the 40th
Annual
Biology Colluquium (ed. R.H. Waring). Oregon State University Press,
Oregon.
321
Sweeney, B.W. (1984) Factors influencing life-history patterns of aquatic
insects. In: The Ecology of Aquatic Insects (eds V.H. Resh and D.M.
Rosenberg), pp.56-100. Praeger Publishers, New York.
Sweeney, B.W. and Vannote, R.L. (1978) Size variation and the distribution
of hemimetabolous aquatic insects: two thermal equilibrium hypotheses.
Science, 200, 444-446.
Sweeney, B.W. and Vannote, R.L. (1981) Ephemerella mayflies of Clay
Creek: bioenergetic and ecological relationships among six coexisting
species. Ecology 62: 1353-1369.
Sweeney, B.W. and Vannote, R.L., and Dodds, P.J. (1986a) The relative
importance of temperature and diet to larval development and adult size in
the winter stonefly. Soyedina carolinensis. Plecoptera: Nemouridae)
Freshwater Biol. 16, 39-48.
Sweeney, B.W. and Vannote, R.L., and Dodds, P.J. (1986b) Effects of
temperature and food quality on growth and development of a mayfly,
Leptophlebia intermedia Can. J. Fish. Aquat. Sci. 43, 12-18.
Tett, P., Gallegos, C., Kelly, M.G., Hornberger, G.M. and Cosby, B.J.
(1978). Relationship among substrate, flow and benthic pigment density in
Mecums Rivers, Virginia. Limnol. Oceanogr. 23, 785-797.
Thibault, M. (1971) Le developpment des Ephemeropteres semivoltines et
univoltins d’un ruisseau de Pays Basque Francais. Doctorat these.
Universite de Paris.
Tilman, D. and Kilman, S.S. (1976) Phosphate and silicate growth and
uptake kinetics of the diatoms Asterionella formosa and Cyclotella
meneghiniana in batch and semicontinuous culture. J. Phycol. 12, 375-383.
Tilzer, M. (1968) Zur Okologie und Besiedlung des hochalpinen
hyporheischen Interstitials im Arlberggebiet (Osterreich). Archiv fur
Hydrobiologie, 65, 253-308.
Tomas, X. and Sabater, S., (1985) The diatom associations in the Llobregat
river and their relations to water quality, Verh. Int. Verein. Limnol., 22,
2348-2352.
Townsend, C.R. (1989) The patch dynamics concept of stream community
ecology. Journal of the North Americam Benthological Society, 8, 36-50.
Townsend, C.R. and Hildrew, A.G. (1994) Species traits in relation to a
habitat template for river systems. Freshwater Biology, 31, 265-276.
Trost, L.M.W. and Berner, L. (1963) The biology of Callibaetis floridanus
Banks (Ephemeroptera: Baetidae). Florida Entomologist, 46, 285-299.
322
Twist, H. Edwards, A.C. and Codd, G.A. (1998) Algal growth responses to
waters of contrasting tributaries of the River Dee, north-east Scotland.
Water Research, 32, 2471-2479.
Van Nieuwenhuyse, E.E. and Jones, J.R. (1996) Phosphorus chlorophyll
relationship in temperate streams and its variation with stream catchment
area. Canadian Journal of Fisheries and Aquatic Sciences, 53, 99-105.
Vannote, R.L. and Sweeney, B.W. (1985) Larval feeding and growth rate of
the stream cranefly Tipula abdominalis in gradients of temperature and
nutrition. Proc. Acad. Nat. Sci. Philadelphia, 137, 119-128.
Vannote, R.L., Minshall, G.W., Cummins, K.W., Sedell, J.R. and Cushing,
C.E. (1980) The river continuum concept. Canadian Journal of Fisheries
and Aquatic Sciences, 37, 130-137.
Vollenweider, R.A. (1968) Scientific fundamentals of eutrophication of
lakes and flowing waters, with particular reference to nitrogen and
phosphorus as factors in eutrophication. Technical report DAS/CSI/68.27,
Environmental Directorate, Organization for Economic Cooperation and
Development (OECD), Paris, 154 p.
Walker, E.D., Olds, E.J. and Merritt, R.W. (1988) Gut content analysis of
Mosquito larvae (Diptera: Culicidae). Using DAPI stain and epifluorescence
microscopy. Entomology Society Of America. 25, 551-554.
Wallace, J.B. and Anderson, N.H. (1996) Habitat, life history and
behavioural adaptations of aquatic insects. In An introduction to the aquatic
insects of North America (ed. R.W. Merritt and K.W. Cummins),pp. 41-73.
Kendall/Hunt, Dubuque, IA.
Wallace, J.B., and Gurtz, M.E. (1986). Response of Baetis mayflies
(Ephemeroptera) to catchment logging. Am. Midl. Nat., 115, 25-41.
Wallace, J.B., and Webster, J.R. (1996) The role of macroinvertebrates in
stream ecosystem function. Annual Review of Entomology, 41, 115-139.
Wallace, J.B., Grubaugh, J.W. and Whiles, M.R. (1996) Biotic indices and
stream ecosystems processes: results from an experimental study. Ecol.
Appl. 6, 140-151.
Walton, S.P., Welch, E.B. and Horner, R.R (1995) Stream periphyton
response to grazing and changes in phosphorus concentration.
Hydrobiologia, 302, 31-46.
Ward, J.V. (1992) Aquatic Insect Ecology. 1. Biology and Habitat. J. Wiley
and Sons, New York, 438p.
Ward, G.M., and Cummins, K.W. (1979) Effects of food quality on growth
of a stream detritivore, Paratendipes albimantus (Meigen) (Diptera:
Chironomidae). Ecology, 60, 57-64.
323
Ward, J.V. and Stanford, J.A. (1979) The ecology of regulated streams.
Plenum Press, New York.
Ward, J.V. and Stanford, J.A. (1984) Thermal responses in the evolutionary
ecology of aquatic insects. Annual Review of Entomology, 27, 97-117.
Warren, P.H. (1989) Spatial and temporal variation in the structure of a
freshwater food web. Oikos, 55, 299-311.
Washington, H.G. (1984) Diversity, biotic and similarity indices: a review
with special relevance to aquatic ecosystems. Wat. Res. 18, 653-694.
Watson, V., Berlind, P., and Bahls, L. (1990) Control of algal standing crop
by P and N in the Clark Fork River. Proc. Clark Fork River Symposium.
Welch, E.B., Quinn, J.M. and Hickey, C.W. (1992) Periphyton biomass
related to point-source enrichment in seven New Zealand streams. Wat. Res.
26, 669-675.
Welch, E.B., Jacoby, J.M., Horner, R.R. and Seeley, M.R. (1988) Nuisance
biomass levels of periphytic alage in streams. Hydrobiologia, 157, 161-168.
Wellnitz, T.A. and Ward, J.V. (1998) Does light intensity modify the effect
mayfly grazers have on periphyton? Freshwater Biology, 39, 135-149.
Wetzel, R.G. (1983) Limnology. 2nd
edn. Philadelphia. W.B. Saunders Co.
WFD CIS Guidance Document No.10 (March 2003) Rivers and Lakes –
Typology, Reference Conditions and Classification Systems. Published by
the Directorate General Environment of the European Commission,
Brussels, ISBN No. 92-894-5614-0, ISSN No. 1725-1087.
Whelan, K.F. (1980) Studies on the life histories, distribution and
production of the Ephemeroptera in selected Irish waters. Ph.D. thesis,
Zoology Dept., University College, Dublin.
Whitney, R.J. (1939) The thermal resistance of mayfly nymphs from ponds
and streams. J. Exp. Biol. 16, 374-385.
Whitton, B.A. (1979) Algae and higher plants as indicators of river
pollution. In Biological Indicators of Water Quality (editors A. James and L.
Evison). John Wiley, Chichester.
Wiederholm, T. (1980) Use of benthos in lake monitoring. Journal of the
Water Pollution Control Federation, 52, 537-547.
Williams, D.D., Williams, N. and Hogg, I. (1995) Life history plasticity of
Nemoura trisinosa (Plecoptera: Nemouridae) along a permanent-temporary
water habitat gradient. Freshwater Biology, 34, 155-63.
Wise, E.J. (1980) Seasonal distribution and life histories of Ephemeroptera
in a Northumbrian river. Freshwater Biology, 10, 101-111.
324
Wittekind, D. (1972) Fluorescence properties of the monoaminoacridines
and some 2-aminoacridine-derivatives, pp. 95-105. In A. A Thaer & M.
Sernetz (eds.), Fluorescence techniques techniques in cell biology. Springer,
New York.
Wong, S.L. and Clarke, B. (1976) Field determination of the critical nutrient
concentratins for Cladaphora in streams. J. Fish. Res. Board Can. 33, 95-
96.
Wood, P.J. and Armitage, P.D. (1997) Biological effects of fine sediment in
the lotic environment. Envir. Mgmt., 21, 203-217.
Wright, J.F. (1995) Development and use of a system for predicting the
macroinvertebrate fauna in flowing waters. Australian Journal of Ecology.
20, 181-197.
Wright, J.F., Sutcliffe, D.W., Furse, M.T. (eds), (2000) Assessing the
Biological Quality of Fresh Waters – RIVPACS and Other Techniques.
Freshwater Biological Association, Ambleside, Cumbria, U.K., 373 PP.
Wuhrmann, K. and Eichenberger, E. (1975) Experiment on the effect of
inorganic enrichment of rivers on periphyton primary production.
Verhandlugen Internationale Vereinigung fur Theoretische und Angewandte
Limnologie, 19, 2028-2034.
Yule, C. (1986) Comparison of the dietary habits of six species of
Dinotoperla (Plecoptera: Gripopterygidae) in Victoria. Aust. J. Mar.
Freshwater Res. 37, 121-127.
Zelinka, M. and Marvan, P. (1961) Zur Prazisierung der biologischen
Klassifikation der Reinheit fliebender Gewasser. Arch. Hydrobiol. 57, 389-
407.
325
Appendices
Appendix 5.1 Maximum, minimum, mean, median and standard deviation values for the physico-chemical parameters examined in the high
status rivers during the study programme, May 2001 to October 2002.
Temperature
°C
DO %
Saturation
pH
pH units
Conductivity
µS/cm
Orthophosphate
mg/l P
TON
mg/l N
Ammonia
mg/l N
Chloride
mg/l Cl
Alkalinity
mg/l
CaCO3
BOD5
Mg/l O2
Colour
Hazen
units
Owengarve River
n 14 14 12 14 14 14 13 13 13 11 14
Minimum 5.3 90.0 7.6 316.0 0.018 0.12 0.01 13.0 116.0 0.3 27.0
Mean 11.7 104.2 7.9 397.7 0.037 0.54 0.03 16.6 167.2 0.8 107.7
Median 11.3 99.8 8.0 387.5 0.038 0.48 0.02 17.0 158.0 0.9 109.0
Maximum 17.0 129.9 8.2 500.0 0.049 1.40 0.07 22.0 252.0 1.4 246.0
Standard
deviation
3.5 12.60 0.2 67.46 0.009 0.31 0.02 2.4 40.79 0.3 60.9
Dunneill River n 14 14 13 14 13 12 13 14 14 12 14
Minimum 6.0 99.0 6.7 73.0 0.012 0.10 0.01 8.0 9.0 0.3 19.0
Mean 11.9 105.2 7.9 260.5 0.032 0.19 0.02 13.7 104.4 0.7 83.7
Median 12.1 104.2 8.1 264.0 0.023 0.15 0.01 14.0 84.0 0.6 47.5
Maximum 16.4 123.0 8.3 588.0 0.112 0.50 0.11 19.0 298.0 1.5 243.0
Standard
deviation
3.3 6.8 0.5 147.8 0.027 0.13 0.03 3.1 76.1 0.4 74.3
Castlebar River n 14 14 13 15 14 14 14 13 14 12 15
Minimum 5.4 95.4 7.3 111.0 0.025 0.10 0.005 16.0 23.0 0.3 44.0
Mean 11.5 101.1 7.7 168.2 0.036 0.16 0.015 21.5 42.9 0.8 109.5
Median 12.2 99.4 7.7 175.0 0.032 0.13 0.010 20.0 36.5 0.6 98.0
Maximum 15.7 110.0 8.2 215.0 0.098 0.3 0.040 30.0 81.0 1.8 228.0
Standard
deviation
3.3 5.0 0.2 32.2 0.019 0.07 0.011 4.37 17.4 0.5 55.6
326
Appendix 5.1 continued. Maximum, minimum, mean, median and standard deviation values for the physico-chemical parameters examined in
the high status rivers during the study programme, May 2001 to October 2002.
Temperature
°C
DO %
Saturation
pH
pH units
Conductivity
µS/cm
Orthophosphate
mg/l P
TON
mg/l N
Ammonia
mg/l N
Chloride
mg/l Cl
Alkalinity
mg/l
CaCO3
BOD5
Mg/l O2
Colour
Hazen
units
Callow Loughs Stream n 14 14 13 14 14 14 14 14 14 11 14
Minimum 5.8 94.0 7.8 173.0 0.018 0.18 0.005 13.0 53.0 0.3 21.0
Mean 11.9 101.8 8.0 286.9 0.035 0.24 0.012 16.9 115.4 0.5 53.3
Median 11.9 99.1 8.0 311.0 0.031 0.20 0.007 17.0 120.0 0.3 33.0
Maximum 18.5 114.0 8.3 348.0 0.082 0.40 0.030 22.0 158.0 0.8 129.0
Standard
deviation
3.9 6.9 0.2 61.13 0.016 0.08 0.008 2.8 34.5 0.3 35.6
Brusna River n 12 12 11 11 11 10 11 12 12 10 12
Minimum 6.1 98.4 7.7 352.0 0.023 0.20 0.005 12.0 136.0 0.3 26.0
Mean 11.8 107.2 8.0 515.0 0.042 0.57 0.011 21.3 230.7 0.6 69.8
Median 12.0 103.0 8.1 521.0 0.038 0.50 0.005 22.0 239.0 0.6 37.5
Maximum 16.8 137.0 8.2 826.0 0.105 1.00 0.030 27.0 296.0 0.9 167.0
Standard
deviation
2.9 11.2 0.2 144.6 0.022 0.27 0.010 3.7 57.7 0.3 54.6
327
Appendix 5.2 Maximum, minimum, mean, median and standard deviation values for the physico-chemical parameters examined in the impacted
rivers during the study programme, May 2001 to October 2002.
Temperature
°C
DO %
Saturation
pH
pH units
Conductivity
µS/cm
Orthophosphate
mg/l P
TON
mg/l N
Ammonia
mg/l N
Chloride
mg/l Cl
Alkalinity
mg/l
CaCO3
BOD5
Mg/l O2
Colour
Hazen
units
Cartron River n 10 10 10 10 11 11 11 10 9 9 10
Minimum 9.4 94.6 5.3 80.0 0.007 0.10 0.01 19.0 6.0 0.3 12.0
Mean 12.5 106.6 6.6 106.3 0.018 0.16 0.01 24.5 8.6 0.8 145.8
Median 12.5 102.9 6.6 101.0 0.014 0.10 0.01 24.0 6.0 0.8 152.5
Maximum 16.1 138.2 8.1 132.0 0.036 0.60 0.02 34.0 24.0 1.3 206.0
Standard
deviation
2.28 12.5 1.0 19.3 0.009 0.16 0.004 4.7 6.0 0.3 61.5
Lough na Corralea Stream n 9 9 9 10 9 10 10 9 9 9 9
Minimum 7.6 87.1 6.1 87.0 0.005 0.10 0.020 16.0 6.0 0.3 42.0
Mean 12.0 97.4 7.1 112.2 0.010 0.14 0.05 16.2 18.2 1.1 75.4
Median 11.5 96.0 7 104.5 0.008 0.1 0.04 15.0 17 0.9 78.0
Maximum 16.5 111.0 8 167.0 0.025 0.4 0.1 23 32.0 3.5 122.0
Standard
deviation
2.9 7.6 0.6 23.6 0.006 0.1 0.03 2.8 8.0 1.0 25.0
Mad River n 11 11 11 12 11 11 11 11 10 11 12
Minimum 7.7 98.6 5.5 42.0 0.005 0.1 0.01 8.0 6.0 0.3 33.0
Mean 11.4 106.8 7.1 54.1 0.023 0.1 0.02 10.1 8.4 0.9 121.7
Median 11.8 103.4 7.3 51.0 0.019 0.1 0.01 10.0 8.0 1.0 130.0
Maximum 16.6 127.0 8.0 74.0 0.061 0.1 0.04 16.0 15.0 1.5 191.0
Standard
deviation
2.7 9.6 0.9 10.1 0.018 0 0.01 2.3 2.8 0.5 48.2
328
Appendix 5.2 continued. Maximum, minimum, mean, median and standard deviation values for the physico-chemical parameters examined in
the impacted rivers during the study programme, May 2001 to October 2002.
Temperature
°C
DO %
Saturation
pH
pH units
Conductivity
µS/cm
Orthophosphate
mg/l P
TON
mg/l N
Ammonia
mg/l N
Chloride
mg/l Cl
Alkalinity
mg/l
CaCO3
BOD5
Mg/l O2
Colour
Hazen
units
Mullaghanoe River n 12 12 12 14 14 14 13 14 14 11 14
Minimum 6.0 97.0 7.7 308.0 0.028 0.69 0.02 17.0 121.0 0.3 14.0
Mean 11.4 110.7 7.8 395.8 0.053 0.99 0.08 18.8 159.2 0.9 43.9
Median 11.8 109.6 7.9 400.5 0.048 0.95 0.07 19.0 162.0 0.9 42.5
Maximum 14.6 141.9 8.1 471.0 0.098 1.40 0.21 21.0 202.0 2.5 74.0
Standard
deviation
2.65 12.8 0.11 50.8 0.018 0.26 0.06 1.37 22.3 0.6 19.8
Robe River n 13 12 11 12 12 12 11 12 13 11 12
Minimum 5.4 90.0 7.7 63.0 0.027 0.50 0.01 17.0 264.0 0.3 12.0
Mean 12.7 107.9 8.0 594.0 0.042 0.83 0.02 19.8 308.3 0.9 54.7
Median 14.1 102.3 8.0 635.0 0.040 0.70 0.01 20.0 304.0 0.9 47.5
Maximum 16.0 158.2 8.4 677.0 0.066 1.60 0.03 22.0 360.0 1.8 138.0
Standard
deviation
3.2 18.2 0.2 169.6 0.010 0.33 0.01 1.48 25.7 0.5 30.3
329
Appendix 5.3 Daily rainfall (mm) for calendar years 2001 and 2002 at Straide (National Grid E: 125780, N: 297380), Co. Mayo which is central to the study area and given
as an indication of wet and dry spells rather than an absolute indication of rainfall for each of the 10 catchments studied. Total rainfall at Straide was 1024 mm in 2001 and
1374 mm in 2004. This compares with the long-term average annual precipitation of 1181 mm at Straide. Long-term evapotranspiration here is 456 mm - 378 mm in summer
and 78 mm in winter giving long-term values of 725 mm annual runoff with 626 mm net winter runoff and 99 mm net summer runoff.
Daily Rainfall (Straide) 2001-2002
0
5
10
15
20
25
30
35
40
1-J
an-0
1
27
-Jan
-01
22
-Feb
-01
20
-Mar
-01
15
-Ap
r-0
1
11
-May
-01
6-J
un
-01
2-J
ul-
01
28
-Ju
l-0
1
23
-Au
g-0
1
18
-Sep
-01
14
-Oct
-01
9-N
ov
-01
5-D
ec-0
1
31
-Dec
-01
23
-Jan
-02
18
-Feb
-02
16
-Mar
-02
11
-Ap
r-0
2
7-M
ay-0
2
2-J
un
-02
28
-Ju
n-0
2
24
-Ju
l-0
2
19
-Au
g-0
2
14
-Sep
-02
10
-Oct
-02
5-N
ov
-02
1-D
ec-0
2
27
-Dec
-02
Da
ily
Ra
infa
ll
( m
m )
330
Appendix 5.4 Comparative descriptive statistics for 99 western rivers over the period 1996-2001 based on 12763 samples at 346 sample sites as
analysed by the EPA Laboratory, Castlebar, Co. Mayo.
Lower Upper 10th 90th
Determinand Valid N Mean Median Min Max Std.Dev. Quartile Quartile percentl percentl
Chloride (mg Cl/l) 12676 44.2 20 1 29320 523.3 17 23 15 27
Colour (Hazen) 12713 74.9 67 5 1250 50.5 39 100 22 142
Conductivity (µS/cm) 12743 466.7 439 7.8 36000 991 287 571 142 633
Hardness (Mg CaCo3/l) 913 179.0 168 28 529 72.6 118 230 99 282
pH 12744 7.9 7.9 5.1 9.3 0.4 7.7 8.1 7.4 8.3
Temperature (°C) 12667 10.63 10.1 0.0 102.3 4.31 7.4 14.0 5.4 16.3
Dissolved Oxygen % 12640 94.9 95 0 232 11.1 89 100 84 106
Ammonia (mgN/l) 12627 0.058 0.028 0.002 81.560 0.858 0.015 0.047 0.009 0.077
BOD (mg O2/l) 12520 1.39 1.2 0.0 240.0 2.43 0.9 1.6 0.6 2.2
Molybdate Reactive Phosphorus
(mg P/l) 12739 0.034 0.018 0.001 33.250 0.322 0.009 0.032 0.008 0.050
Oxidised Nitrogen (mg N/l) 12692 0.918 0.687 0.009 41.512 0.923 0.271 1.286 0.100 2.024
Nitrite (mgN/l) 1298 0.007 0.004 0.000 0.200 0.012 0.002 0.007 0.002 0.011
Suspended Solids (mg/l) 1368 6.5 5 0 208 11.2 3 6 1 11
Copper (mg/l) 1332 0.005 0.001 0.001 1.200 0.043 0.001 0.002 0.001 0.007
Zinc (mg/l) 1331 0.044 0.020 0.001 13.600 0.444 0.004 0.025 0.002 0.025
3
31
Ap
pen
dix
5.5
Box
and W
hisk
er plo
ts of th
e seasonal v
ariation in
the M
argalef’s in
dex
(d) (S
pecies rich
ness) o
ver fiv
e seasons acro
ss the ten
rivers.
Margalef's index (d)
Ow
eng
arv
e
0 1 2 3 4 5 6 7
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02C
astleb
ar
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Bru
sna
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Du
nn
eill
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Callo
w L
ou
gh
s Strea
m
0 1 2 3 4 5 6 7
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Cartro
n
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Lou
gh
na C
orra
lea
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Rob
e
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Mu
llag
han
oe
0 1 2 3 4 5 6 7
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Mad
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
±1
.96
*S
td. D
ev.
±1
.00
*S
td. D
ev.
Mean
3
32
Ap
pen
dix
5.6
Box
and W
hisk
er plo
ts of th
e seasonal v
ariation in
the to
tal num
ber o
f taxa p
er metre sq
uare o
ver fiv
e season
s across th
e ten riv
ers.
Total number of taxa (S)
Ow
eng
arv
e
0
15
30
45
60
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
stleba
r
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Bru
sna
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Du
nn
eill
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
llow
Lo
ug
hs S
tream
0
15
30
45
60
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
rtron
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02L
ou
gh
na
Co
rralea
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ro
be
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Mu
llag
ha
no
e
0
15
30
45
60
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ma
d
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
±1
.96
*S
td. D
ev.
±1
.00
*S
td. D
ev.
Mean
3
33
Ap
pen
dix
5.7
Box
and W
hisk
er plo
ts of th
e seasonal v
ariation in
the to
tal num
ber o
f indiv
iduals p
er metre sq
uare o
ver fiv
e seasons acro
ss the ten
rivers.
Total number of individuals (N)
Ow
en
ga
rv
e
-4000 0
4000
8000
12000
16000
20000
24000
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02C
astle
ba
r
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Bru
sna
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Du
nn
eill
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
llow
Lou
gh
s Str
ea
m
-4000 0
4000
8000
12000
16000
20000
24000
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
rtr
on
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Lou
gh
na
Corra
lea
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Rob
e
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Mu
llag
ha
noe
-4000 0
4000
8000
12000
16000
20000
24000
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ma
d
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
±1
.96
*S
td. D
ev.
±1
.00
*S
td. D
ev.
Mean
3
34
Ap
pen
dix
5.8
Box
and W
hisk
er plo
ts of th
e seasonal v
ariation in
the P
ercentag
e EP
T p
er metre sq
uare o
ver fiv
e seasons acro
ss the ten
rivers.
Percentage EPT
Ow
eng
arv
e
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
stleba
r
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Bru
sna
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Du
nn
eill
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
llow
Lo
ug
hs S
tream
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
rtron
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Lo
ug
h n
a C
orra
lea
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ro
be
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Mu
llag
ha
no
e
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ma
d
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
±1
.96
*S
td. D
ev.
±1
.00
*S
td. D
ev.
Mean
3
35
Ap
pen
dix
5.9
Box
and W
hisk
er plo
ts of th
e seasonal v
ariation in
the S
han
non-W
iener in
dex
over fiv
e seasons acro
ss the ten
rivers.
Shannon-Wiener diversity index
Ow
eng
arv
e
0.2
0.8
1.4
2.0
2.6
3.2
3.8
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
stleba
r
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Bru
sna
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Du
nn
eill
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
llow
Lo
ug
hs S
tream
0.2
0.8
1.4
2.0
2.6
3.2
3.8
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
rtron
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Lo
ug
h n
a C
orra
lea
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ro
be
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Mu
llag
ha
no
e
0.2
0.8
1.4
2.0
2.6
3.2
3.8
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ma
d
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
±1
.96
*S
td. D
ev.
±1
.00
*S
td. D
ev.
Mean
3
36
Ap
pen
dix
5.1
0 B
ox
and W
hisk
er plo
ts show
ing th
e seasonal v
ariation in
Sim
pso
n’s in
dex
over fiv
e seasons acro
ss the ten
river sites.
Simpsons index
Ow
eng
arv
e
0.1
0.4
0.7
1.0
1.3
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Castleb
ar
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Bru
sna
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Du
nn
eill
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Callo
w L
ou
gh
s Strea
m
0.1
0.4
0.7
1.0
1.3
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Cartro
n
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Lou
gh
na C
orra
lea
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Rob
e
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Mu
llag
han
oe
0.1
0.4
0.7
1.0
1.3
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Mad
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
±1
.96
*S
td. D
ev.
±1
.00
*S
td. D
ev.
Mean
3
37
Ap
pen
dix
5.1
1 B
ox
and W
hisk
er plo
ts of th
e seasonal v
ariation in
Pielo
u’s ev
enness in
dex
(J’) over fiv
e seasons acro
ss the ten
rivers.
Pielou's evenness index
Ow
eng
arv
e
0.1
0.4
0.7
1.0
1.3
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
stleba
r
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Bru
sna
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Du
nn
eill
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
llow
Lo
ug
hs S
tream
0.1
0.4
0.7
1.0
1.3
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ca
rtron
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Lo
ug
h n
a C
orra
lea
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ro
be
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Mu
llag
ha
no
e
0.1
0.4
0.7
1.0
1.3
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
Ma
d
Summer '01
Autumn '01
Spring '02
Summer '02
Autumn '02
±1
.96
*S
td. D
ev.
±1
.00
*S
td. D
ev.
Mean
338
Appendix 5.12 Comparison of the diversity indices and matrices using ANOVA between the high status and impacted sites during the study period.
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p =0.239S
ha
nn
on
-Wie
ner i
nd
ex
(H
')
1.0
1.4
1.8
2.2
2.6
3.0
3.4
High status rivers Impacted rivers
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Ma
rga
lef'
s in
dex
(d
)
1
2
3
4
5
6
High status rivers Impacted rivers
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p = 1
Pie
lou
's e
ven
nes
s in
dex
(J)
0.35
0.45
0.55
0.65
0.75
0.85
0.95
High status rivers Impacted rivers
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p = 0.239
Sim
pso
ns
ind
ex
0.55
0.65
0.75
0.85
0.95
1.05
High status rivers Impacted rivers
339
Appendix 5.12 continued. Comparison of the diversity indices and matrices using ANOVA between the high status and impacted sites during
the study period.
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Tota
l n
um
ber
of
taxa (
S)
5
10
15
20
25
30
35
40
45
50
High status rivers Impacted rivers
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p = 0.2390
To
tal
nu
mb
er o
f in
div
idu
als
(N
)
-6000
-4000
-2000
0
2000
4000
6000
8000
10000
12000
High status rivers Impacted rivers
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Per
cen
tage
EP
T
0.1
0.2
0.3
0.4
0.5
0.6
High status rivers Impacted rivers
340
Appendix 5.13 Comparison of the biotic indices AQEM between the high status and impacted sites during the study period.
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p = 0.009
Du
tch
Sa
pro
bic
In
dex
-0.25
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Ger
man
Sap
rob
ic I
nd
ex
1.4
1.5
1.6
1.7
1.8
1.9
2.0
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Czech
Sap
rob
ic I
nd
ex
0.9
1.1
1.3
1.5
1.7
1.9
2.1
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Dan
ish
Str
eam
Fau
na I
nd
ex (
DS
FI)
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
Impacted sites High status sites
341
Appendix 5.13 Continued: Comparison of the biotic indices AQEM between the high status and impacted sites during the study period.
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Average S
core p
er T
axon
(A
SP
T)
4.4
4.8
5.2
5.6
6.0
6.4
6.8
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Ind
ice B
ioti
co E
steso
(IB
E)
5
6
7
8
9
10
11
12
13
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Bel
gia
n B
ioti
c In
dex
(B
BI)
4.5
5.5
6.5
7.5
8.5
9.5
10.5
Impacted sites High status sites
342
Appendix 5.14 Comparison of the microhabitat preferences calculated using AQEM between the high status and impacted sites during the study period.
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Pel
al
mic
roh
ab
itat
typ
e (m
ud
; gra
in s
ize
< 0
.063)
-10
-5
0
5
10
15
20
25
30
35
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Lit
hal
mic
roh
ab
itat
typ
e (g
rain
siz
e >
2cm
)
10
20
30
40
50
60
70
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Ph
yta
l m
icro
hab
itat
typ
e (a
lgae,
moss
es, m
acr
op
hyte
s)
5
10
15
20
25
30
35
40
45
Impacted sites High status
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Ak
al
mic
roh
ab
itat
typ
e (g
rain
siz
e 0.2
-2cm
)
-6
0
6
12
18
24
30
Impacted sites High status sites
343
Appendix 5.15 Comparison of the feeding types calculated using AQEM between the high status and impacted sites during the study period.
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Fee
din
g t
yp
e (M
iner
s)
-1.5
-0.5
0.5
1.5
2.5
3.5
4.5
5.5
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Fee
din
g t
yp
e (g
ath
erer
s/co
llec
tors
)
0
10
20
30
40
50
60
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Fee
din
g t
yp
e (a
ctiv
e fi
lter
fee
der
s)
-4
-2
0
2
4
6
8
10
12
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001F
eed
ing t
yp
e (G
raze
rs a
nd
scr
ap
ers)
0
20
40
60
80
100
Impacted sites High status
344
Appendix 5.15 Comparison of the feeding types calculated using AQEM between the high status and impacted sites during the study period.
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p = 0.002
Fee
din
g t
yp
e (P
red
ato
rs)
-6
-2
2
6
10
14
18
22
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p < 0.001
Fee
din
g t
yp
e (P
ara
site
s)
-2
-1
0
1
2
3
4
5
Impacted sites High status sites
±1.96*Std. Dev.
±1.00*Std. Dev.
Mean
p = 0.002
Fee
din
g t
yp
e (P
ass
ive
filt
er f
eed
ers)
-20
-10
0
10
20
30
40
50
Impacted sites High status sites
345
Appendix 5.16 Substrate fraction weights (grams) for the high status and impacted rivers sampled in March 2003.
High status
rivers
Owengarve River Dunneill River
Castlebar River Callow Loughs Stream Brusna River
Sieve (mm) Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
0.03 2 13 13 1 1 1 1 0 0 1 0 1 1 0 1
0.063 7 46 24 5 2 3 2 2 1 2 3 5 2 2 2
0.125 25 188 84 11 6 10 2 3 1 2 10 20 3 2 5
0.25 42 202 68 35 15 41 6 6 1 6 12 43 5 2 11
0.50 61 180 35 127 51 138 38 29 5 38 13 119 30 6 37
1 109 160 68 278 188 190 113 56 26 113 23 195 197 12 126
2 69 68 32 180 332 93 131 38 48 131 20 110 115 3 57
4 240 154 132 34 224 388 408 152 187 408 14 250 229 30 144
8 286 215 208 38 672 430 440 199 380 440 44 135 180 24 68
16 589 511 717 11 537 1015 629 573 650 629 380 9 548 155 193
32 735 415 443 1383 1676 962 1114 1314 962 1327 550 570 471 430
64 246 1345 4475 1049 3555 2535 4648 3261 2535 1605 2003 2166 2528 2870
128 2337 5214 701 2873 701 2650 3412 1172 850
256 2636
Total 2411 1737 5478 5638 12310 7540 5968 9693 5874 5968 3451 6090 7458 4407 4794
Impacted
Rivers
Cartron River Lough na Corralea Stream Mad River Robe River Mullaghanoe River
Sieve (mm) Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
Sample 1
Weight (g)
Sample 2
Weight (g)
Sample 3
Weight (g)
0.03 1 1 0 0 2 1 0 1 1 2 1 1 1 1 0
0.063 3 9 1 1 6 3 1 3 1 6 5 5 2 4 2
0.125 8 24 2 1 10 8 1 6 3 11 13 11 6 14 6
0.25 9 44 21 12 21 16 4 9 9 24 30 20 16 33 10
0.50 14 66 136 9 41 47 18 37 43 43 64 28 72 32 10
1 8 114 165 19 82 99 64 172 149 52 155 41 155 31 9
2 13 88 171 119 84 178 53 183 115 36 134 36 95 25 19
4 129 526 739 220 388 117 357 729 410 45 492 146 259 149 45
8 213 620 868 728 727 413 261 712 138 93 541 269 195 225 49
16 1085 1400 1613 699 665 140 548 1287 518 2798 2071 1143 265 939 86
32 1690 866 641 3258 1213 520 1023 965 1178 2143 1532 1591 342 1600 227
64 4411 4835 1819 726 2963 791 1385 2061 1146 1276 1523 3487 1120 4468 2675
128 969 2735 1899 1572
256
Total 8553 11328 6176 5792 8101 4175 3715 6165 3711 6529 6561 6778 2528 7521 3138
346
Appendix 5.17 Percentage frequency of each substrate fraction (by weight) for the five high status sites studied in March 2003.
High status
rivers
Owengarve River Dunneill River Castlebar River
Sieve (mm)
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
0.03 0.08 0.75 0.24 0.36 0.02 0.01 0.01 0.01 0.02 0.00 0.00 0.01
0.063 0.29 2.65 0.44 1.13 0.09 0.02 0.04 0.05 0.03 0.02 0.02 0.02
0.125 1.04 10.82 1.53 4.46 0.20 0.05 0.13 0.13 0.03 0.03 0.02 0.03
0.25 1.74 11.63 1.24 4.87 0.62 0.12 0.54 0.43 0.10 0.06 0.02 0.06
0.50 2.53 10.36 0.64 4.51 2.25 0.41 1.83 1.50 0.64 0.30 0.09 0.34
1 4.52 9.21 1.24 4.99 4.93 1.53 2.52 2.99 1.89 0.58 0.44 0.97
2 2.86 3.91 0.58 2.45 3.19 2.70 1.23 2.37 2.20 0.39 0.82 1.13
4 9.95 8.87 2.41 7.08 0.60 1.82 5.15 2.52 6.84 1.57 3.18 3.86
8 11.86 12.38 3.80 9.35 0.67 5.46 5.70 3.95 7.37 2.05 6.47 5.30
16 24.43 29.42 13.08 22.31 0.20 4.36 13.46 6.01 10.54 5.91 11.07 9.17
32 30.49 0 7.58 12.69 7.86 11.23 22.23 13.77 16.12 11.49 22.37 16.66
64 10.20 0 24.55 11.58 79.37 8.52 47.15 45.01 42.48 47.95 55.52 48.65
128 0 0 42.67 14.22 0 42.36 0 14.12 11.75 29.64 0 13.80
256 0 0 0 0 0 21.41 0 7.14 0 0 0 0
Total 100 100 100 100 100 100 100 100 100 100 100 100
High status
rivers
Callow Loughs Stream Brusna River
Sieve (mm)
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
0.03 0.02 0.00 0.02 0.01 0.01 0.00 0.01 0.01
0.063 0.03 0.09 0.08 0.07 0.03 0.05 0.04 0.04
0.125 0.03 0.29 0.33 0.22 0.04 0.05 0.06 0.05
0.25 0.10 0.35 0.71 0.38 0.07 0.05 0.11 0.08
0.50 0.64 0.38 1.95 0.99 0.40 0.14 0.44 0.33
1 1.89 0.67 3.02 1.86 2.64 0.27 1.85 1.59
2 2.20 0.58 1.81 1.53 1.54 0.07 0.93 0.85
4 6.84 0.41 4.11 3.78 3.07 0.68 2.25 2.00
8 7.37 1.27 2.22 3.62 2.41 0.53 1.46 1.47
16 10.54 11.01 0.15 7.23 7.35 3.52 4.96 5.28
32 16.12 38.45 9.03 21.20 7.64 10.69 9.10 9.14
64 42.48 46.51 32.89 40.62 29.04 57.37 48.76 45.06
128 11.75 0 43.51 18.42 45.75 26.60 30.03 34.12
256 0 0 0 0 0 0 0 0
Total 100 100 100 100 100 100 100 100
347
Appendix 5.18 Percentage frequency of each substrate fraction (by weight) for the five impacted sites studied in March 2003.
Impacted
rivers
Cartron River Lough na Corralea Stream Mad River
Sieve (mm)
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
0.03 0.01 0.01 0.00 0.03 0.02 0.00 0.02 0.02 0.00 0.02 0.03 0.02
0.063 0.04 0.08 0.02 0.08 0.07 0.02 0.07 0.05 0.03 0.05 0.03 0.04
0.125 0.09 0.21 0.03 0.20 0.19 0.02 0.12 0.11 0.03 0.10 0.08 0.07
0.25 0.11 0.39 0.34 0.63 0.38 0.21 0.26 0.28 0.11 0.15 0.24 0.17
0.50 0.16 0.58 2.20 1.11 1.13 0.16 0.51 0.60 0.48 0.60 1.16 0.75
1 0.09 1.01 2.67 1.25 2.37 0.33 1.01 1.24 1.72 2.79 4.02 2.84
2 0.15 0.78 2.77 3.64 4.26 2.05 1.04 2.45 1.43 2.97 3.10 2.50
4 1.51 4.64 11.97 6.69 2.80 3.80 4.79 3.80 9.61 11.82 11.05 10.83
8 2.49 5.47 14.05 12.20 9.89 12.57 8.97 10.48 7.03 11.55 3.71 7.43
16 12.69 12.36 26.12 14.83 9.82 12.07 8.21 10.03 14.75 20.88 13.96 16.53
32 19.76 7.64 10.38 26.91 12.46 56.25 14.97 27.89 27.54 15.65 31.75 24.98
64 51.57 42.68 29.45 26.52 18.95 12.53 36.58 22.69 37.28 33.43 30.89 33.87
128 11.33 24.14 0 5.91 37.65 0.00 23.44 20.36 0 0 0 0
256 0 0 0 0 0 0 0 0 0 0 0 0
Total 100 100 100 100 100 100 100 100 100 100 100 100
Impacted
rivers
Robe River Mullaghanoe River
Sieve (mm)
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
Sample 1
%
Sample 2
%
Sample 3
%
Mean
%
0.03 0.03 0.02 0.01 0.02 0.04 0.01 0.00 0.02
0.063 0.09 0.08 0.07 0.08 0.08 0.05 0.06 0.06
0.125 0.17 0.20 0.16 0.18 0.24 0.19 0.19 0.21
0.25 0.37 0.46 0.30 0.38 0.63 0.44 0.32 0.46
0.50 0.66 0.98 0.41 0.68 2.85 0.43 0.32 1.20
1 0.80 2.36 0.60 1.25 6.13 0.41 0.29 2.28
2 0.55 2.04 0.53 1.04 3.76 0.33 0.61 1.57
4 0.69 7.50 2.15 3.45 10.25 1.98 1.43 4.55
8 1.42 8.25 3.97 4.55 7.71 2.99 1.56 4.09
16 42.85 31.57 16.86 30.43 10.48 12.49 2.74 8.57
32 32.82 23.35 23.47 26.55 13.53 21.27 7.23 14.01
64 19.54 23.21 51.45 31.40 44.30 59.41 85.25 62.99
128 0 0 0 0 0 0 0 0
256 0 0 0 0 0 0 0 0
Total 100 100 100 100 100 100 100 100
348
Appendix 5.19 Cumulative percentage frequency distribution of the substrate in the high status rivers measured during the sediment analysis programme.
(A) Cumulative percentage frequency distribution of substrate in the Owengarve
River, March 2003. Sample 1: Sample 2: Sample 3: Mean
(B) Cumulative percentage frequency distribution of substrate in the Dunneill
River, March 2003. Sample 1: Sample 2: Sample 3: Mean
(C) Cumulative percentage frequency distribution of substrate in the Castlebar
River, March 2003. Sample 1: Sample 2: Sample 3: Mean
(D) Cumulative percentage frequency distribution of substrate in Callow Loughs
Stream, March 2003. Sample 1: Sample 2: Sample 3: Mean
0
10
20
30
40
50
60
70
80
90
100
-1.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log10
Sieve size (mm)
Percen
tage f
iner
0
10
20
30
40
50
60
70
80
90
100-1
.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log10
Sieve size (mm)
Percen
tage f
iner
0
10
20
30
40
50
60
70
80
90
100
-1.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log 10 Sieve size (mm)
Percen
tage f
iner
0
10
20
30
40
50
60
70
80
90
100
-1.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log10
Sieve size (mm)
Percen
tage f
iner
A B
C
D
349
Appendix 5.19 Cumulative percentage frequency distribution of the substrate in the Brusna River (high status river) measured during the sediment
analysis programme.
(E) Cumulative percentage frequency distribution of substrate in the Brusna
River, March 2003. Sample 1: Sample 2: Sample 3: Mean
0
10
20
30
40
50
60
70
80
90
100
-1.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log10
Sieve size (mm)
Percen
tage f
iner
E
350
Appendix 5.20 Cumulative percentage frequency distribution of the substrate in the impacted rivers measured during the sediment analysis programme.
(A) Cumulative percentage frequency distribution of substrate in the Cartron
River, March 2003. Sample 1: Sample 2: Sample 3: Mean
(B) Cumulative percentage frequency distribution of substrate in Lough na Corralea
Stream, March 2003. Sample 1: Sample 2: Sample 3: Mean
(C) Cumulative percentage frequency distribution of substrate in the Mad
River, March 2003. Sample 1: Sample 2: Sample 3: Mean
(D) Cumulative percentage frequency distribution of substrate in the Robe River,
March 2003. Sample 1: Sample 2: Sample 3: Mean
0
10
20
30
40
50
60
70
80
90
100-1
.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log10
Sieve size (mm)
Percen
tage f
iner
0
10
20
30
40
50
60
70
80
90
100
-1.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log10
Sieve size (mm)
Percen
tage f
iner
0
10
20
30
40
50
60
70
80
90
100
-1.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log10
Sieve size (mm)
Percen
tage f
iner
0
10
20
30
40
50
60
70
80
90
100
-1.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log10
Sieve size (mm)
Percen
tage f
iner
A
B
C D
351
Appendix 5.20 Cumulative percentage frequency distribution of the substrate in the Mullaghanoe River (impacted rivers) measured during the sediment
analysis programme.
(E) Cumulative percentage frequency distribution of substrate in the Mullaghanoe
River, March 2003. Sample 1: Sample 2: Sample 3: Mean
0
10
20
30
40
50
60
70
80
90
100
-1.5
-1.2
-0.9
-0.6
-0.3 0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
Log10
Sieve size (mm)
Percen
tage f
iner
E
353
Recommended